Technical Assistance Consultant’s Report

Contract No. 123137-S86665

TA-8789 REG: CAREC Knowledge Sharing and Services in Transport and Transport Facilitation

Road Asset Management

Road Map

Sector Specialist - Road Asset Management

July 2019

Prepared by: Serge Cartier van Dissel

This consultant’s report does not necessarily reflect the views of ADB or the Government concerned, and
ADB and the Government cannot be held liable for its contents.
EXECUTIVE SUMMARY
The ADB was recently requested by the Government of Tajikistan to support the development of a Road
Asset Management System (RAMS) for the identification of maintenance needs in the 14,220 km of roads
managed by the Ministry of Transport (MOT) and to increase the efficiency and effectiveness of maintenance
interventions. This document is the result of an exploratory one-week mission to Dushanbe and a subsequent
one-week analysis of existing studies and documents. This document looks at the development of a RAMS,
including the three main aspects of data collection for the road network, data management to facilitate the
access to the collected data, and the subsequent data analysis to determine maintenance needs and
prioritize interventions. This document further looks at the integration of the RAMS into the wider context
of road asset management, including the institutional framework within which the RAMS would be placed,
the planning procedures into which the RAMS would need to be integrated, the financing sources that are
available for funding the identified priority interventions, and the implementation modalities that may be
applied to carry out the funded interventions. These aspects are all discussed in the following paragraphs.

DATA COLLECTION
The collection of data for the road network in Tajikistan is quite limited. Data is collected by the GUSADs in
annual assessments carried out together with the traffic police by means of visual assessments and manual
traffic counts, with only limited use of equipment-based measurements. Data is not georeferenced to the
location where it was collected. The survey vehicle provided to MOT under the CARs-2 project is only used
for surveying completed road works, and the data output is not considered suitable for a RAMS. The data
collection proposed by the CARs-2 consultant is based on old soviet GOST standards and is considered
inappropriate for a modern RAMS. It is recommended to provide a new survey vehicle with modular survey
equipment that may expanded in the future and that is able to collect and combine GPS, distance, roughness
and video data, allowing this to be imported directly into the RAMS in a georeferenced database format. This
would be complemented by GPS-enabled handheld equipment to collect the remaining data in the road
network. The set of data to be collected should initially be kept to a minimum, reducing the time and effort
spent each year. The data would focus on the main properties and functional condition of the road pavement,
bridges and tunnels, complemented with traffic data.

Roads Bridges Tunnels Traffic

• GPS track • GPS location • GPS location • Vehicle counts by type
• Length (chainage) • Length • Length • Axle loads
• Roughness • Main characteristics • Main characteristics
• Surface condition • Structural condition • Structural condition
• Video • Photographs/Video • Photographs/Video

DATA MANAGEMENT
Collected data is currently not being stored in a systematic manner and is spread over different electronic
documents and paper reports. There is no central repository for the data, and a lot of data remains with the
GUSADs and is unavailable at MOT. A Microsoft Access database was prepared in 2008 but is currently in
disuse. A new RAMS database is currently being developed under the CARs-2 project, but this is a tailor-made
system with basic queries and reporting functions hard-wired into the system. The ability of this system store
and manage the georeferenced data, allowing for detailed analysis and exporting to other software, appears
to be limited. It is also unclear how future technical support will be provided. It is recommended that a more
flexible database system be introduced that allows for proper integration with the data collection by the
survey vehicle, as well as with other data analysis software. It is strongly recommended to use commercial
off-the-shelf software with strong technical support services. This should include both the database and the
GIS mapping functions.

DATA ANALYSIS
In its simplest form, data analysis may include a basic statistical yearbook describing the main characteristics
of the road network each year. Although MOT publishes such a yearbook, this only looks at transport statistics

i
and does not include road network statistics. It is strongly recommended to expand the yearbook to also
include road network statistics prepared using the RAMS. Data analysis goes beyond simple statistics,
however, and aims to identify maintenance needs and prioritize budget allocations to those roads that have
the highest benefits for road users. Under the Cars-2 project, the consultant is currently developing a set of
algorithms to identify such needs. These are tailor-made algorithms that have not been tested elsewhere,
and their suitability may therefore be questioned. They are also hard programmed into the RAMS software,
complicating future adjustments. Rather than using such unproven algorithms, it is recommended to initially
use off-the-shelf software such as HDM-4, that has a proved track record in many countries, including in the
CAREC region. This may be used to directly define (multi)annual work programmes using data exported from
the RAMS. HDM-4 may also be used to prepare a strategy analysis based on the overall make-up of the road
network and the expected budget levels. This will help determine optimal treatment types and identify
priority road types, which may then form the basis for the planning module in the RAMS.

INSTITUTIONAL FRAMEWORK
The international and republican roads as well as many of the local roads are managed by MOT through the
Economic Analysis & Forecasting Department (EAFD) and the Road Construction and Maintenance
Department (RCMD) in Dushanbe, the 6 Regional Offices and the 62 State Institutes for Road Maintenance
(GUSADs). Several extensive studies have been carried out on institutional reform, including the creation of
a Road Fund and a Road Authority, and the commercialization and possible privatization of the GUSADs.
Rather than dealing with such extensive reforms, this document focuses on the creation of a specific RAMS
unit under MOT. This would be created directly under one of the deputy ministers, allowing it to work
together with the different departments involved in the road sector. The RAMS unit would initially have
approximately 6 staff positions and would need office space as well as equipment and software for data
collection, data management and data analysis. It would also require an annual budget allocation to carry
out the data collection and operate the RAMS. Since the creation of such a unit will take some time, initially
a task force will be created with staff from the RCMD and EAFD. This task force may initially work together
with the RAMS consultant and receive training, with some members later transferring to the RAMS unit.

PLANNING
Currently planning follows former Soviet procedures. Maintenance needs are identified by the GUSADs
together with the traffic police and forwarded to the Regional Offices. Budget requests are prepared on the
basis of unit rates and are later combined into a consolidated budget request by MOT. This does not reflect
all needs and is reduced to approximately double the expected budget allocation. The approved budget is
spread amongst the GUSADs on the basis of the network size managed by each of them. The introduction of
the RAMS will introduce a major change in this process. Although data collection will still be carried out
locally, the identification of maintenance needs and the preparation of the budget request will be carried out
centrally. The allocation of funding will be targeted at the prioritized roads and will no longer be evenly
spread between the GUSADs on the basis of the road length managed by them. To ensure that the benefits
of a RAMS will be achieved, a detailed assessment of the current planning procedures is required, identifying
how the RAMS may be integrated into this process and identifying any specific obstacles that may need to
be removed. The RAMS consultant will furthermore be required to prepare multiannual plans for at least two
consecutive years, which will form the basis for the annual budget requests for road maintenance.

FINANCING
For 2019 the amount received for road maintenance from the Republican Budget was $7.2 million, which is
less than one-quarter of the lowest estimated actual need (varying from $30-$110 million per year). Although
the budget allocation has doubled in the past 10 years in Tajikistan Somoni terms, the budget allocation has
remained more or less the same in United States Dollar terms. Additional funding is urgently required to
ensure the proper maintenance of the roads that have been improved over the past decade. This will require
the allocation of revenue from road user charges, possibly channelling these through a Road Fund. Tajikistan
currently has a Road User Tax which generated $42 million in 2018, but this is actually a business tax that is
used for financing road investments and the government has been planning to abolish this tax for several
years. Road tolling may form an additional source of revenue for high volume roads and a simple tolling
system managed by MOT is planned to be piloted under the CAREC Corridors 2, 3, and 5 (Obigarm–Nurobod)

ii
Road Project. A fuel excise tax is a common road user charge in other countries and although Tajikistan
currently has such a tax, its revenue is not allocated to the road sector. An additional fuel excise tax of
$0.05/litre for petrol and $0.03/litre for diesel could generate $19 million per year. A vehicle tax is collected
in Tajikistan with a reported revenue of $20 million in 2018. However, this revenue is collected by the
municipal authorities and allocated to the municipal budgets, and is therefore considered less suitable for
financing a central Road Fund. Vehicle import fees including customs duties and excise taxes on vehicles may
form a more suitable source of revenue since they are collected centrally. Overloading fines are currently
collected by the traffic police, but the fines and resulting revenue are very low. Increasing the fines may be a
suitable means of generating more revenue from those road users that cause the most damage to the roads.
Introducing such road user charges and linking these to a Road Fund will be an important step in increasing
the amount of road maintenance funding. This will require that the revenue from these charges is to a certain
extent earmarked for the Road Fund, ensuring higher predictability of funding and facilitating multiannual
road maintenance planning and contracting.

IMPLEMENTATION
Although there are many registered contractors licensed to undertake road construction and maintenance
work in Tajikistan, half of these are state-owned companies including the GUSADs. The involvement of private
sector contractors in road maintenance is limited to some heavier periodic maintenance works and
performance-based maintenance (PBM) contracts under development partner projects. Most road
maintenance currently involves routine and winter maintenance that is carried out by the GUSADs using basic
equipment, whereby half the budget is spent on staff costs. The introduction of a RAMS is expected to shift
the focus towards the implementation of periodic maintenance, which will likely take up approximately half
the road maintenance budget. This requires the development of the necessary implementation skills and
contracting modalities. Although it is possible to expand the capacities of the GUSADs to also carry out
periodic maintenance, it is recommended to instead develop this capacity with the private sector, creating a
more flexible system that is better able to respond to demand. It is also recommended to further build on
the existing experience with PBM contracts to include periodic maintenance. The PBM contracts will also
require improvement regarding the applied response times, performance standards and penalty deductions
in order to create proper incentives that motivate contractors to improve their performance.

ROAD MAP
For improving the road asset management in Tajikistan, with particular emphasis on the development of a
road asset management system and improved road maintenance funding, the road map at the end of this
document is proposed. This foresees a series of different steps to be undertaken over the next 5 years, from
2020 to the end of 2024. For 2020 a number of basic activities are foreseen, including the procurement of
data collection and data management equipment and software, the identification of a RAMS taskforce in
MOT, the detailed analysis of current planning systems and the review of the ongoing PBM pilots. By the end
of 2021, it is expected that data will be collected for at least 1,000 km of international roads and that the
RAMS and related GIS mapping will be developed, allowing an initial road network analysis to be carried out
and a first multiannual maintenance programme to be developed. In this year a detailed study of funding
needs and user charges will also be completed. In 2022 the RAMS unit will be formally created, which will
receive an operational budget to continue data collection and management, forming the basis for an updated
multiannual maintenance programme. This year will also see the introduction of some road user charges, as
well as the procurement of new PBM contracts. By the end of 2023 the data collection will be completed for
all international and republican roads, allowing a full road network analysis to take place as basis for the
multiannual road maintenance programme. Road network statistics will start to be published on an annual
basis, and the set of road user charges will be expanded to include the first tolling pilots. By the end of 2024
the collection of condition data will be repeated for the majority of international roads, allowing MOT to
update their maintenance planning. With the introduction of the road user charges, road maintenance
financing will be increased to at least 0.25% of GDP.

iii
CONTENTS
Executive Summary....................................................................................................................... i
1. Introduction ............................................................................................................................ 1
2. Road Network ......................................................................................................................... 1
3. Road Asset Management System ............................................................................................. 2
Data collection .............................................................................................................................. 2
Data management ........................................................................................................................ 7
Data analysis ................................................................................................................................. 9
4. Institutional framework ........................................................................................................ 11
5. Planning................................................................................................................................ 13
6. Financing .............................................................................................................................. 14
Maintenance Budget .................................................................................................................. 14
Maintenance Needs.................................................................................................................... 15
Road Fund ................................................................................................................................... 17
Road User Tax ............................................................................................................................. 18
Tolling ......................................................................................................................................... 18
Fuel Excise Tax ............................................................................................................................ 21
Vehicle Tax .................................................................................................................................. 21
Overloading fines and fees ......................................................................................................... 22
Vehicle Import Duties ................................................................................................................. 24
7. Implementation .................................................................................................................... 24
GUSADs ....................................................................................................................................... 25
Performance-Based Maintenance (PBM) ................................................................................... 26
8. Road Map ............................................................................................................................. 28
Road Asset Management System (RAMS) .................................................................................. 28
Institutional framework .............................................................................................................. 28
Planning ...................................................................................................................................... 29
Financing ..................................................................................................................................... 29
Implementation .......................................................................................................................... 29

iv
TABLES
Table 1 Road network managed by MOT .......................................................................................... 2
Table 2 Estimated road conditions .................................................................................................... 2
Table 3 Proposed datasets for the CARs-2 RAMS.............................................................................. 4
Table 4 Recommended minimum data collection............................................................................. 5
Table 5 Recommended road network survey equipment ................................................................. 6
Table 6 Standard queries proposed by CARs-2 RAMS consultant ..................................................... 8
Table 7 MOT road maintenance budgets (million) .......................................................................... 15
Table 8 Road maintenance funding as percentage of GDP ............................................................. 15
Table 9 Estimated long-term maintenance needs .......................................................................... 16
Table 10 Toll rates on Dushanbe-Khujand-Chanak toll road (TJS)..................................................... 19
Table 11 IRS regional offices and related road sections .................................................................... 19
Table 12 Estimated annual toll revenue Dushanbe-Khujand-Chanak ............................................... 20
Table 13 CAREC corridors in Tajikistan .............................................................................................. 20
Table 14 Fuel usage in Tajikistan (2015) ............................................................................................ 21
Table 15 Number of registered vehicles (December 2015)............................................................... 22
Table 16 Vehicle Tax rates by vehicle type ........................................................................................ 22
Table 17 Weight limits and applied fines .......................................................................................... 23
Table 18 Customs duties.................................................................................................................... 24
Table 19 Number of GUSADs and road length (km) by Regional Office ............................................ 25
Table 20 Road maintenance budget allocation ................................................................................. 25
Table 21 Costs of first phase PBM pilots (2013-2016)....................................................................... 26
Table 22 Costs of second phase PBM pilots (2018-2020).................................................................. 26
Table 23 Road Map ............................................................................................................................ 31

FIGURES
Figure 1 RAMS and the wider context of road asset management .................................................... 1
Figure 2 DRIMS system ....................................................................................................................... 3
Figure 3 Road survey vehicle provided under CARs-2 ........................................................................ 4
Figure 4 Example of a modular system from ROMDAS ...................................................................... 6
Figure 5 Organizational structure of MOT ........................................................................................ 11
Figure 6 Proposed RAMS unit ........................................................................................................... 12
Figure 7 Expected future road network condition under different budget scenarios ..................... 16
Figure 8 Toll stations in Varzob and Istaravshan .............................................................................. 19
Figure 9 Hissar and Kyrgyz border weigh stations ............................................................................ 23
Figure 10 Warzob weighing station .................................................................................................... 23

v
ABBREVIATIONS
ADB Asian Development Bank
BT Bituminous
CAREC Central Asia Regional Economic Cooperation
DRIMS Dynamic Response Intelligent Monitoring System
EAFD Economic Analysis and Forecasting Department
EAR Euro
EBRD European Bank for Reconstruction and Development
ER Earthen
GDP Gross Domestic Product
GIS Geographic Information System
GOST State Standard of the Soviet Union (ГОСТ)
GPS Global Positioning System
GR Gravel
GUSAD State Institute for the Maintenance of Roads (ГУСАД)
HDM Highway Design and Management
HIMS Highway Information Management system
HIS Highway Information System
IMF International Monetary Fund
IRI International Roughness Index
IRS Innovative Road Solutions
IT Information Technology
JICA Japan International Cooperation Agency
JSC Joint Stock Company
LAN Local Area Network
LLC Limited Liability Company
LTD Limited
MOF Ministry of Finance
MOT Ministry of Transport
MS Microsoft
NHA National Highway Authority
OPRC Output- and Performance-based Road Contract
PBM Performance-Based Maintenance
PC Personal Computer
PDF Portable Document Format
QGIS Quantum Geographic Information System
RAMS Road Asset Management System
RCMD Road Construction and Maintenance Department
RMA Road Maintenance Account
ROMDAS Road Measurement Data Acquisition System
TJS Tajiki Somoni
TOR Terms of Reference
TRACECA Transport Corridor Europe-Caucasus-Asia
US United States
USD United States Dollar
VAT Value Added Tax
WAN Wide Area Network

vi
1. INTRODUCTION
1. This document looks at the current approaches to road maintenance and road asset management in
Tajikistan. It is based on a one-week mission to Dushanbe to discuss with MOT and development partners,
followed by a one-week review of existing studies and documents. This document aims to identify the main
areas for improvement and the steps to be taken in doing so over the next few years. It looks at the
introduction of a Road Asset Management System (RAMS), focusing both on the RAMS itself as well as the
wider context of road asset management in terms of the institutional framework, the planning and budgeting
procedures, the sources of financing, and the implementation modalities. These are all described in the
following sections, identifying the current status and the main areas for improvement. This document focuses
on the introduction of a RAMS and its integration into planning procedures, and on improving the financing of
road maintenance. Other aspects of road asset management related to institutional reform and the
involvement of the private sector in road maintenance are dealt with in other studies and are not repeated in
detail here. The document concludes with a summary of the main steps to be undertaken towards better road
asset management over the next few years, which are reflected in a Road Map presented at the end of this
document.

Figure 1 RAMS and the wider context of road asset management

2. ROAD NETWORK
2. Tajikistan is a very mountainous country with a land area of 143,000 km2. It is divided into 5 regions1,
which are subdivided into 62 districts (including 4 districts belonging to the capital Dushanbe). Tajikistan has a
road network of approximately 26,000 km, of which 14,220 km are managed by the Ministry of Transport
(MOT). This includes 3,348 km of international roads (24%), 2,129 km of republican roads (15%) and 8,744 km
of local roads (61%). According to the Law of the Republic of Tajikistan about Roads and Road Activities (2002
and amended in 2016), local governments are formally the owners and managers of local roads. However,
because the local governments lack the financial means and technical expertise for maintaining all the local
roads situated in their jurisdiction, important local roads are maintained and managed by MOT. The result is
that nearly two-thirds of the network managed by MOT consists of local roads. The network also includes 2,234
bridges with a combined length of nearly 43 km, as well as nearly 20,000 culverts with a combined length of
over 212 km. In addition, there are over 17 km of tunnels in the road network.

1 Dushanbe, the Districts of Republican Subordination, Sughd Region, Khatlon Region, Gorno-Badakhshan Autonomous Region

1
 Table 1 Road network managed by MOT

Roads (km) Bridges (m) Culverts (m)

Category Bituminous
Asphalt Gravel Unpaved Total # Length # Length
gravel
International 1,762 1,231 353 2 3,348 515 15,391 5,282 73,951
Republican 938 728 427 36 2,129 331 7,757 2,900 32,661
Local 1,865 3,807 1,919 1,152 8,744 1,388 19,565 11,502 106,068
Total 4,565 5,766 2,699 1,190 14,220 2,234 42,713 19,684 212,680
32% 41% 19% 8% 100%
Source: MOT 2019

3. Tajikistan distinguishes 5 technical road classes (I, II, III, IV and Y). Approximately 85% of the network is in
class III or IV, with very few higher-class roads and the remainder classified as underclass Y. Just under three-
quarters of the MOT road network is paved (90% of international roads), of which just under half has an asphalt
concrete surface and the rest has a surface of gravel mixed with bitumen. Of the remaining roads, most have
a gravel surface with just under one-tenth of the MOT road network having an earthen surface (these are
almost exclusively local roads).

4. Most of the network is in poor condition. The 2011 Transport Sector Master Plan estimated that over 80%
of the network was beyond simple repair, with more than half the road length having a roughness greater than
7 m/km. The average speed in the network was estimated to be only 30 km/h. The poor road conditions are
especially prevalent in the local roads, with international roads generally in far better condition.

Table 2 Estimated road conditions

Condition International roads Republican roads Local roads
Good 35% 28% 8%
Fair 50% 24% 16%
Poor 15% 48% 76%
Source: MOT, 2016

5. The Road Construction & Maintenance Department and the Economic Analysis & Forecasting Department
under MOT are responsible for the road network. Implementation of routine and winter maintenance is largely
carried out through 6 Regional Offices and 62 GUSADs at district level set up as State Institutes under the Road
Construction & Maintenance Department. Apart from the Regional Offices and GUSADs, there is also a
Transport Design Institute set up as a State Unitary Enterprise under MOT.

3. ROAD ASSET MANAGEMENT SYSTEM

6. A RAMS generally involves data collection regarding the road network and its condition, data
management to store the data and make it easily accessible, and data analysis to use the data as the basis for
planning and monitoring. These three aspects of a RAMS are described in the following sections.

DATA COLLECTION

7. Data collection by MOT is currently very limited. Data on road length, surface type, number of lanes and
structures (number and length of bridges and tunnels, number of culverts) are collected every year, but these
are not georeferenced using GPS data, nor are they systematically stored in any database. Data on road
condition and maintenance needs is collected annually in January by means of visual inspections carried out
by the GUSADs together with the traffic police, resulting in a report “for Republican Commands of Police
Inspector”. Traffic data is collected by means of manual traffic counts on an annual basis (sometimes more
frequently). However, this data is only available for certain locations.

8. Systematic data collection using survey equipment is not carried out, although such surveys have been
carried out in the past under development partner funded projects. In 2008 a road inventory and condition
survey was carried out for approximately 4,000 km of roads under the ADB technical assistance “Strengthening

2
Implementation of the Road Maintenance Financing System” (TA 4294 TAJ). However, this data was not kept
up-to-date and is currently no longer in use.

9. More recently, JICA has been supporting basic data collection in 4 regions2. Use has been made of the
Dynamic Response Intelligent Monitoring System (DRIMS), which collects data on roughness using a class III
accelerometer attached above the rear wheel of a vehicle and combines this with data from a basic handheld
GPS device, with data recorded in a laptop PC which calculates IRI-values for 100m sections. The accelerometer
is calibrated by driving the vehicle over two standard humps at a predetermined speed. A JICA funded TA has
been promoting this equipment with the GUSADs in the regions and has collected roughness data for
approximately 1,300 km of roads in 2013, 2014, 2015 and 2016 (Gissar and Kurgontebba regions). Video data
is also collected using a dashcam, which is later used to identify crack ratios and pothole ranks at 100m
intervals. Results are visualized using Google Earth.

Figure 2 DRIMS system

10. Even more recently, the World Bank’s second phase of the Central Asia Road Links project (CARs-2) has
provided various data collection equipment, including 41 fixed traffic counting stations. The traffic counting
stations have a 24h video camera system to monitor the traffic flow. The traffic counting stations are linked to
a server in Moscow, but MOT is aiming to have a server in country to manage the traffic data. Only 7 of the
traffic counting stations have so far been programmed and are operational. It is not yet clear where the
different traffic counting stations will be located and how their data will be stored and analysed. The traffic
counting stations are operated by the GUSADs under MOT.

11. Under the CARs-2 project, a survey vehicle was also purchased. The Trassa survey vehicle is made in Russia
and provided by a Tajikistani company called “LLC Shayid”. It collects roughness data using a laser profilometer.
The data is not georeferenced, however, and is exported in PDF format (reportedly to avoid the data from
being manipulated). This makes the data less suitable for use in a RAMS as it cannot be easily imported or
linked to GPS coordinates and GIS maps. The survey vehicle is currently with the Transport Design Institute
under MOT and is used mainly to check the quality of construction works after completion. Network surveys
have not been carried out.

2 Mainly Gissar and Kurgontebba, and to a more limited extent in Sugd and Kulyab.

3
 Figure 3 Road survey vehicle provided under CARs-2

12. Under the CARs-2 project, support is also being provided to the development of a road asset management
system for MOT. This includes an electronic road mapping module and a road database module. For the
mapping module, the Kazakh consultant (Prof. Krasikov) recommends obtaining the required GIS data
commercially from the Kazakh company Brief Research Group. It would appear that this data is obtained from
existing maps and satellite images, rather than collecting this directly using GPS devices. Although this is
acceptable for much information (e.g. district boundaries, rivers and lakes), for the road specific data it is not
considered to be sufficiently accurate and it is strongly recommended to acquire this data directly.

13. The CARs-2 consultant also identifies a set of 16 main tables or datasets that will form the database as
listed below. For each of the datasets, data collection forms are provided indicating the data to be collected.
The data to be collected for the different roads is limited to general data on the start and end point, the names
of districts crossed, the road length, and the number of bridges, overpasses and culverts. Further additional
data includes the number of lanes, the presence of medians, the surface type, the traffic volume and data on
the year of construction, rehabilitation and major repair. Geometric data is also collected with respect to the
carriageway width, shoulder width, side drain width and right-of-way by means of measurements carried out
for the cross section in a typical location. This data is collected for the road (link) as a whole, rather than
collecting GPS coordinates and using this as the basis for sectioning the roads into uniform segments.

Table 3 Proposed datasets for the CARs-2 RAMS

1. List of roads and general data

2. Additional data on the roads
3. Geometric parameters of the roads (cross section)
4. Visual assessment of the strength of the pavement and road conditions
5. Evaluation of the pavement roughness (bump integrator/profilometer)
6. Location of road accidents in the past 3 years
7. Traffic intensity
8. Bus stop characteristics and condition
9. Car park and petrol station characteristics and condition
10. Culverts
11. Pavement structure and strength
12. Bridges, overpasses and bypasses
13. Technical condition of bridges, overpasses and their visual images
14. Repair works
15. Line diagrams of longitudinal profile, road surface condition and surface, traffic, intersections, etc.
16. Status of tunnels

14. For the road condition data, the CARs-2 consultant has defined four condition categories (excellent, good,
fair, poor) based on a combination of a visual assessment of the pavement condition (depending on the

4
percentage of cracking, potholes and other pavement defects), the pavement roughness (IRI) and the roadside
condition (depending on shoulder condition and vegetation control). It is not exactly clear, however, how the
different assessment criteria relate to each other (for instance, a road with uncut vegetation would rate as
poor, irrespective of the pavement condition).

15. For traffic data, the CARs-2 consultant has provided formulas for converting the results of three-hour
traffic counts (09:00-12:00 or 15:00-18:00) to average daily traffic counts. The data collection for bus stops,
car parks and petrol stations is quite extensive, including aspects such as the number of petrol pumps, fire
safety status, distance from residential area, etc.). The amount of data required for structures is even more
extensive. However, in the end the data appears to only be used to determine the condition category of the
road element, without necessarily providing useful data for further analysis.

16. The data collection and assessments appear to be based on current GOST standards dating from Soviet
times, rather than actually identifying the data needs for a road asset management system. Based on the initial
review of the data collection proposed by the CARs-2 consultant, it is expected that this will not be suitable for
a proper road asset management system since the data is collected at too high a level to allow for appropriate
planning. At the same time, a lot of data is collected that is not a priority (e.g. regarding the number of petrol
pumps and the siltation of culvert pipes). Geographic location data appears to be lacking.

17. For the road network data collection, it is recommended to use data collection equipment as much as
possible. This increases the accuracy and speed with which the data can be collected, reducing the cost. The
data to be collected should be kept to a minimum initially, increasing this over time as MOT gains experience
with the RAMS and additional needs are identified. The data collection should initially focus on the road
properties and functional condition, tunnels and bridges, and traffic data. It is recommended to collect the
following data initially:

Table 4 Recommended minimum data collection

Roads Bridges Tunnels Traffic

• GPS track • GPS location • GPS location • Vehicle counts by type
• Length (chainage) • Length • Length • Axle loads
• Roughness • Main characteristics • Main characteristics
• Surface condition • Structural condition • Structural condition
• Video • Photographs/Video • Photographs/Video

18. The GPS tracks and locations can be collected using GPS equipment installed on a survey vehicle,
complemented by handheld GPS devices for the structures. Data on road, bridge and tunnel lengths is best
collected using an accurate odometer installed on a survey vehicle (measuring tape may be used for smaller
structures). Accurate roughness data is best collected using a laser profilometer. However, these devices are
easily damaged on unpaved and poor condition roads, for which a bump integrator may be used (this may be
installed in a different vehicle for use on poor roads)3. The video data may be collected using a video camera
installed on the survey vehicle. All roughness, chainage and video data needs to be linked to the GPS data to
facilitate further processing. The post-processing of the video data allows the location of different structures
to be identified (GPS and chainage), the surface types and number of lanes to be determined, and the surface
condition to be assessed. This is often preferable to doing this assessment during the survey, as it avoids
surveyor fatigue and results can easily be reviewed and checked. For the roads, most of the data can be
collected in one single drive-over survey with different equipment installed on a single survey vehicle. For the
bridges and tunnels, it will be necessary to stop to collect additional data. It is recommended to carry this out
in a separate survey, collecting basic data on location, length, type, materials and condition from the drive-
over survey first. A specific bridge and tunnel survey (possibly also including other structures such as culverts)
may be carried out at a later date.

3Alternatively use may be made of simple accelerometer-based equipment such as the DRIMS or smartphone-based apps such as
RoadLab or RoadRoid.

5
19. To facilitate objective data collection that can be used in the RAMS, it is recommended to procure
additional data collection equipment. Preferably this would be a new vehicle4 with modular equipment
installed that may be expanded as required. The equipment should include an accurate odometer for
measuring distance (chainage), an accurate GPS device for measuring location, a laser profiler for measuring
roughness (this may be complemented with a bump integrator for use in poor condition roads), and at least
one video camera for collecting video data for the road. The number of video cameras may be increased to
allow better coverage of the roadway, especially the roadside and the opposite lane5. It is important that the
data from the different survey equipment is linked together, ensuring that roughness data and video data is
linked to the chainage data and to the GPS data. This allows the data to be viewed together during
postprocessing, allowing further analysis of the data. It is further recommended that the different modules be
procured from a single supplier that provides proper support during installation and operation of the
equipment and related software. An example of such a modular system is given in Figure 4. This will need to
be complemented with the necessary data collection software and data postprocessing software.

Table 5 Recommended road network survey equipment

• Survey vehicle (4x4)

• Odometer
• GPS device
• Laser profilometer (roughness good/fair roads)
• Bump integrator (roughness poor roads)
• Video camera(s)
Figure 4 Example of a modular system from ROMDAS

20. For the traffic data, use may be made of the fixed traffic counting stations provided under the CARs-2
project. It is important that suitable locations for this equipment be properly identified, ensuring that the data

4This would preferably be a 4x4 vehicle that does not get easily damaged by use in poor roads.
5 This would include one camera looking straight forward to collect data on the travelling lane, one camera aimed at the roadside to
collect data on roadside features, and one aimed at the opposite lane. A fourth camera aimed directly down at the pavement to collect
video data on pavement damage is not considered necessary at this stage as this considers considerable postprocessing.

6
reflects the traffic volumes in key parts of the road network. It is recommended to complement the fixed traffic
counting stations by a number of sets of portable traffic counters that may be applied in different locations to
collect additional traffic data. Moving traffic counts using the collected video data may form an alternative
where traffic data is not readily available.

21. Data on axle loads can be collected from weigh stations. Here it is preferable to use axle loads rather than
gross weight. This requires that data from existing weigh stations is collected and made accessible to the RAMS.
With the limited number of weigh stations in place, and especially stations that measure axle loads, it is
recommended to increase the number of fixed weigh stations in priority locations. It is further recommended
to purchase a number of sets of portable weigh stations to collect axle load data for other parts of the road
network.

22. The data collection should initially focus on the international roads, and can later be expanded to the
republican roads. It is not recommended to include the local roads at this stage. As part of the support to the
RAMS development under the CAREC Corridors 2, 5, and 6 (Dushanbe–Kurgonteppa) Road Project, it is
recommended to make the RAMS consultants responsible for collecting data for at least 1,000 km of
international roads using the survey equipment to be procured under the project. This will allow the data
collection to be tested, along with the data management software and the data analysis. If the budget and
time allows, this may be expanded to include (most of) the remaining international road network and part of
the republican road network. Data for any remaining roads will need to be collected by MOT using the same
equipment. In discussion with MOT, it was agreed that the survey equipment should remain under MOT and
its GUSADs, rather than placing it with the Technical Design Institute. The survey vehicle would be managed
by the RAMS unit to be formed under MOT in coordination with GUSADs.

DATA MANAGEMENT

23. The collected road network data needs to be properly managed, storing this in a database and making it
easily accessible for further analysis and use. This is currently not happening. Collected data is available in
different reports and Excel or Word files, making it difficult to locate and access. There is no central repository
of the collected data, with much data remaining at the GUSAD level and only summaries being sent to MOT.

24. A database was developed in 2008 under ADB TA 4294 TAJ “Strengthening Implementation of the Road
Maintenance Financing System. This Highway Information System (HIS) was a Microsoft Access database with
a Russian Interface developed by Intercontinental Consultants and Technocrats Pvt Limited from India in
association with the Technical Design Institute and Alternative Solutions LLC from Tajikistan. Road network
inventory and condition data was collected for approximately 4,000 km of roads. However, the database has
not been updated since 2008 and the database is no longer in use (a copy of the database could not be found).
Despite the Russian interface, it is expected that the database was not considered to be easily accessible due
to the use of MS Access software, with which people are not very familiar. MS Access is also not very suitable
for large datasets and would not have been appropriate for managing the data of the entire road network
managed by MOT. However, further effort should be made to obtain a copy of the database in order that the
contents and structure may be taken into account when preparing a new RAMS database.

25. Under the CARs-2 project, a contract was signed in 2016 with an individual consultant from Kazakhstan
(Prof. Krasikov) to support the development of a road asset management system. The consultant is responsible
for preparing the Terms of Reference (TOR) for the staff of the proposed RAMS unit and the TOR for the
procurement of RAMS software, hardware, equipment and consultancy services. The consultant is also
responsible for identifying the training needs (including identifying potential training providers and preparing
a training plan), for preparing an action plan for annual data collection, for preparing an action plan for
introducing and further developing the RAMS, for preparing a user manual for RAMS and GIS management and
data collection, for preparing procedures for monitoring the quality of collected data, for preparing the
procedures for annual reporting, and for preparing an initial 5-year rolling plan for maintenance works under
different budget scenarios.

7
26. The proposed CARs-2 RAMS database will reportedly include a start page, an information page
(description, user instructions, links to normative documents and data forms for data queries), an analysis page
(with different analytical tasks to select from), and a data input page. A number of standard queries have been
identified that will be programmed into the database. These will allow standard reports to be prepared, either
for the network as a whole or for selected roads. The data will be presented in table format, and in some cases
visualized on a map using colour coding (e.g. white-excellent, green-good, yellow-fair, red-poor).

Table 6 Standard queries proposed by CARs-2 RAMS consultant

• Road list with general information

• Pavement condition and surface type
• Traffic volumes
• Location and number of traffic accidents
• Location and condition of bridges, overpasses and tunnels
• Location and condition of culverts
• Location and condition of bus stops, car parks and petrol stations
• Pavement strength
• Repair works

27. The CARs-2 road database is understood to be custom built, although it is not clear what software will be
used. From the description of its functionality, it is expected to be very rigid in its operation, with several
standardized queries hard-programmed into the software. This will likely limit its functionality, as additional
queries will require reprogramming of the software. It is also not clear to which extent the consultant will be
responsible for providing support during operation, allowing problems and desired changes to be addressed.
This is a common weakness of tailor-made software, where the user is dependent on the consultants for future
support.

28. It is also not clear to which degree the data from the CARs-2 database may be exported to a spreadsheet
format (e.g. MS Excel or comma separated values) or to another standard database format (e.g. MS Access)
for further analysis or use in other software (e.g. HDM-4). The export function is an important element of any
RAMS database and needs to be provided. This also requires that the database can identify homogeneous road
sections in which the main road variables remain more or less constant (condition, traffic, surface type, class,
etc.), allowing the data to be exported by homogeneous sections for use in HDM-4 or other pavement
management systems.

29. The database and related mapping functions should be widely accessible, allowing different users to
access and use the data. The main database should be located in MOT (or alternatively in a cloud managed by
MOT). It is recommended to allocate the responsibility for the database to a specific RAMS unit under MOT
(although the actual equipment with the database may be located elsewhere). MOT staff in the different
departments, regional offices and GUSADs should have access to the database through a Local Area Network
(LAN) or Wide Area Network (WAN). It is recommended to also make the database available to other users
through an internet connection. To facilitate its use, the database will therefore require a simple web interface
so that the data may be accessed without the need to install any specific software.

30. The web interface would need to provide access to all data, whereby the access may be restricted to
specific data types according to the type of user (e.g. RAMS unit and other staff involved in maintenance
planning would have full access; other departments, regional offices and GUSADs may have more limited
access; and the general public would have access only to general data). Similar restrictions would exist for
determining whether users are able to add and change data or not. The web interface would allow a number
of standard reports and maps to be prepared, as well as allowing specific filtering criteria to be used. The web
interface would allow filtered data for specific roads or according to specific criteria to be printed and exported
to a spreadsheet or PDF format. The web interface would include a mapping function, allowing filtering results
to be presented visually.

8
31. At this stage the underlying database will be relatively simple, and it is recommended to develop a custom
tailor-made database. Commercial off-the-shelf databases are often much more complex, allowing a far
greater variety of data to be stored and analysed. Having such an extended functionality without the
corresponding data would be a waste at this stage. The expanded functionality of such commercial database
software also comes at a price, with the cost of ROMDAS HIMS software, for instance, ranging from $80,000-
$250,000 for the basic version on a single computer, to $500,000-$1,250,000 for a version with a web interface.

32. For the mapping function, it is recommended to use commercial GIS software. ArcGIS software is
commonly used for this function. However, some countries make use of freeware such as QGIS, which has
extensive functionality and good support services. Although ArcGIS has more extensive functionality, an
assessment will need to be made to determine to which degree this additional functionality is required at this
stage, against the benefits of using software such as QGIS that is freely accessible to everybody (compared to
ArcGIS that requires expensive licenses). It is also possible to make use of the database functionality of GIS
software, although this will require the high-end licenses of ArcGIS. This is also less easily linked to a web
interface.

DATA ANALYSIS

33. In its simplest form, data analysis may take the form of a simple annual yearbook or statistical report of
the road network and traffic information. Although MOT prepares an annual statistical yearbook that is
published on its website, this only includes transport statistics (goods and passenger transport by volume and
volume-distance, as well as vehicle ownership). Statistics on the road network are not included in the annual
statistical yearbook. Once a road database is in place and assuming the data is kept up-to-date, basic statistics
for the road network can be easily prepared. This can identify road lengths by class, surface type, condition
and traffic category. This may then be compared to previous years (as is currently done for the transport data),
allowing trends to be visualized. The required statistics for the road network can be programmed to be a
standard output of the RAMS. It is strongly recommended that MOT include such road network statistics in the
annual statistical yearbook.

34. Although road statistics are useful, data analysis goes beyond that. The most common form of RAMS data
analysis refers to pavement modelling, predicting how pavements of different roads will deteriorate over time
(as a result of age, climate, traffic, etc.) and determining how available budgets are best allocated to maximise
the benefits for road users and the country as a whole. The data analysis will not only need to predict how the
pavement of a specific road will change over time, but also how this will change for all the roads in the network,
and what the influence will be of allocating the available budget to interventions in certain roads (improving
the conditions of those roads) while leaving other roads without investments. This type of modelling is quite
complicated, with many variables that need to be properly programmed and calibrated to the context of a
country (type of vehicles using the road, type of construction materials, type of climate and effect on pavement
deterioration, etc.). Such pavement management systems generally allow the parameters to be adjusted so
that the pavement deterioration predictions better fit the actual situation (this requires actual pavement
conditions to be monitored over time as a basis for comparison).

35. Such a data analysis was carried out in 2008 using the data from the Highway Information System. Use
was made of the Highway Design and Management software (HDM-4). HDM-4 is a commercial off-the-shelf
software that is used in many countries. A strategy analysis was carried out to determine the budgetary
requirements and to identify how available funding could be best spent6. Medium-term rolling programmes
were also prepared, identifying specific road sections to be prioritized and the maintenance treatments to be
applied. HDM-4 has also been used more recently under TA-8945 TAJ, carrying out a strategy analysis based
on available data.

6 In a strategy analysis, the road network is divided into homogeneous segments with similar characteristics (e.g. class, surface type,
condition, traffic volume) and the allocation to the different homogeneous categories is optimized. This also allows the budget
requirements to be identified to address all maintenance needs (unconstrained), and allows the impact on pavement performance of
different budget scenarios to be determined (constrained).

9
36. The CARs-2 RAMS consultant is currently developing pavement deterioration algorithms and economic
optimization algorithms specifically for Tajikistan. These are reportedly based on the algorithms of HDM-4,
which have been simplified to reduce the data needs and calibration requirements. However, the resulting
algorithms have not been extensively tested, and it is unclear how well they will perform in the context of
Tajikistan. With a lack of historical pavement performance data, it will also be impossible to test the algorithms
and compare them to actual deterioration. The use of such tailor-made algorithms therefore introduces
significant risks. This is even more the case since the algorithms appear to be programmed into the planning
module, with limited scope for adjusting and calibrating them. It also means that there is a lack of support
services and experience from users in other countries that can be used as a basis for testing and calibrating
the system in Tajikistan. For a country just starting to introduce road asset management, this is not a
recommended approach. Instead it is strongly recommended to use an off-the-shelf software with an
extensive user base and a wide range of experience, and for which adequate support services exist. This may
at a later stage be replaced by a tailor-made software if considered preferable (this was the approach used in
Belarus, for example).

37. For now, it is recommended to use HDM-4, since a number of countries in the CAREC region are already
successfully making use of this software. This software has been applied in a large number of countries
throughout the world, and it is relatively easy to find experienced trainers and developers that can support
MOT in the calibration and use of the software. The main disadvantage of the software is that it is only available
in English, French and Spanish, and that there is no Russian interface. However, this may be overcome by
providing a clear User Manual in Russian or Tajik.

Pakistan has been using HDM-4 as a pavement management tool since the beginning of this century. The
National Highway Authority (NHA) that is responsible for national highways, manages a RAMS through its
Road Asset Management Directorate (RAMD). HDM-4 is used to prioritize interventions and to determine
the required annual road maintenance budget. Funding is provided from a Road Maintenance Account
(RMA).

In Georgia the use of HDM-4 is more recent. Since 2008 the Roads Department has been developing a RAMS
with development partner support, collecting road data for most of the main network. HDM-4 was
introduced to carry out the data analysis and identify priority interventions and required budgets. HDM-4 is
currently used to prepare the annual maintenance program each year. This is subsequently adjusted on the
basis of additional prioritization criteria before submission for budget approval.

38. Although HDM-4 perhaps has a higher data requirement than the algorithms being developed by the
CARs-2 consultant, much of the data required is related to the calibration of the algorithms (e.g. Vehicle
Operating Costs, Travel Time Costs and Accident Costs), and will need to be entered only once. In the CARs-2
algorithms, it would appear that the variables are hard-programmed into the software – although this reduces
the need for data, it makes it very difficult to make adjustments to the algorithms at a later date and to properly
calibrate them once more historical data on pavement deterioration has been collected. A limited calibration
of HDM-4 was recently carried out under TA-8945 TAJ “CAREC Corridors 2.5 and 6 Road project (Dushanbe-
Kurgonteppa)”. Further calibration may be done at a later stage after road condition data has been collected
for a number of years.

39. The operation of the HDM-4 software will be carried out by the proposed RAMS unit under MOT in
Dushanbe. The GUSADs and other departments and units under MOT will not be directly involved in the use
of HDM-4, and will instead use the outcomes of the HDM-4 analysis in preparing the (multi-)annual plans and
budget requests. To make the results of the HDM-4 analysis available to these users, they will need to be
imported back into the database and linked to the GIS mapping, allowing data to be presented in both tabular
and map form.

10
4. INSTITUTIONAL FRAMEWORK
40. The Ministry of Transport (MOT) is responsible for the management of most of the road network in
Tajikistan, including all the international and republican roads as well as a significant portion of the local roads.
The remaining local roads are managed by the local governments of the 62 districts. MOT has 6 departments,
of which the Road Construction & Maintenance Department (RCMD) and the Economic Analysis & Forecasting
Department (EAFD) are directly involved in road maintenance (the International Relations Department is
involved in development partner funded road projects). Under the RCMD, there are 6 Regional Offices and 62
GUSADs (State Institutes for the Maintenance of Roads7) that are responsible for the road network and that
are set up as State Institutes under MOT.

Figure 5 Organizational structure of MOT

MINISTRY OF TRANSPORT
4 Human Resources Unit
3 Internal Audit Unit
2
Legal Unit

12 15 10 12 8 7
Economic Analysis & Road Construction International Finance &
Inland Transport Administration
Forecasting & Maintenance Relations Accounting
Department Department
Department Department Department Department
6 5 4 6 4
Maintenance Foreign Investment Road Secretariat
Analysis &
Planning & Cooperation Transport
Forecasting Unit
Evaluation Unit Unit Unit
4 5 5 5
Foreign Countries Rail
Road Construction
IT Unit and International Transport
& Safety Unit
Relations Unit Unit

6 Regional Offices

62 GUSADs

41. In the past there have been several studies calling for a reform of MOT and its GUSADs. The most detailed
of these was prepared by EBRD and identified the need for i) the establishment of a second generation Road
Fund to provide secure and stable funding for the road sector, ii) the establishment of an autonomous and
independent Road Authority to take over the road network management function from MOT, iii) the
commercialization and possible privatization of the GUSADs to improve their efficiency and competitiveness,
and iv) the reorganization of MOT to focus on policy formulation, regulations and performance monitoring of
the Road Authority as well as the contracting sector. Other studies call for more limited reforms, including the
reduction of the number of GUSADs and the number of staff they have, and rearranging certain functions
under MOT.

42. This document does not go into the details of the wider institutional reform of MOT and its GUSADs and
instead focuses on the operation of the road asset management system by MOT. The RAMS, once developed,
will need to be operated and maintained by somebody. Ideally this would involve the creation of a specific
RAMS unit that would be responsible for managing and operating the road database and carrying out the data
analysis using HDM-4 and other software. It would also be responsible for coordinating the regular collection

7 ГУСАД - Государственное учреждение по содержанию автомобильных дорог

11
and updating of data, either directly or through a third party. Apart from the creation of the RAMS unit, this
document does not propose any further institutional reform related to road network management.

43. Road maintenance planning is currently jointly undertaken by the Maintenance Planning & Evaluation
Unit under RCMD and the Analysis & Forecasting and IT Units under RAFD. Because this function is spread over
different departments, it was agreed in discussions with MOT that the RAMS unit would ideally be located as
a separate unit directly under one of the deputy ministers. This would be a new unit with approximately 6 staff
members. The option of creating the RAMS unit as an autonomous State Unitary Enterprise or State Institute
was also discussed, but it was decided that this was not appropriate at this stage. The proposed location of the
RAMS unit is indicated in the figure below.

Figure 6 Proposed RAMS unit

MINISTRY OF TRANSPORT
4
Human Resources Unit
6 3
RAMS Unit Internal Audit Unit
2
Legal Unit

12 15 10 12 8 7
Economic Analysis & Road Construction International Finance &
Inland Transport Administration
Forecasting & Maintenance Relations Accounting
Department Department
Department Department Department Department
6 5 4 6 4
Maintenance Foreign Investment Road Secretariat
Analysis &
Planning & Cooperation Transport
Forecasting Unit
Evaluation Unit Unit Unit
4 5 5 5
Foreign Countries Rail
Road Construction
IT Unit and International Transport
& Safety Unit
Relations Unit Unit

6 Regional Offices

62 GUSADs

44. The creation of the RAMS unit will take some time, as it will require the approval of an additional budget
for the staff positions (some staff may be relocated from existing positions, but new positions will likely need
to be added to the current structure of MOT) and for its operation. It is expected that the creation of the RAMS
unit may be realized in 2021. Until then, it is recommended to work with a task force including 2-3 staff
members from each of the three existing units involved in road maintenance planning. These staff members
will work as counterparts to the ADB RAMS consultant, supporting the development of the RAMS and receiving
on-the-job and formal training. Some of these staff members will later transfer to the new RAMS unit. MOT
will provide a list of names of the proposed staff members in the next few months.

45. Apart from funding for the staff positions, the RAMS unit will require equipment including desktop
computers, software licenses (e.g. HDM-4, ArcGIS), a server (although the existing MOT server may be used
instead), a colour printer for reports, and a large-scale colour plotter for preparing the maps. It is
recommended that this equipment be provided as part of the support from ADB. The total cost is estimated
to be in the order of $50,000-$60,000.

46. As mentioned earlier, MOT will also need new road survey equipment and portable traffic counting
stations. It is recommended that this also be provided under the ADB support, with the survey vehicle initially
used by the ADB RAMS consultant for the collection of data in 1,000 km or more of international and republican
roads. This will allow the vehicle and RAMS to be tested and will allow MOT staff to be given on-the-job

12
training. The survey vehicle will remain with the RAMS unit under MOT for collection of data in the remainder
of the international and republican road network. To this end, the ADB RAMS consultant will be responsible
for preparing a data collection programme together with the members of the RAMS task force. The costs of
the survey vehicle are estimated to be in the order of $200,000 (including vehicle). The RAMS unit will need to
receive training in the operation of the survey equipment.

47. The members of the task force will receive training in the operation and use of the database. It is
recommended that the MOT staff (especially the IT staff) be fully involved in the development of the database
and the linkage with the GIS software. This will provide them with on-the-job training and make them aware
of the working of the software. For the development of the software, the ADB RAMS consultant will be required
to work with a local company which may provide continued support in the future.

48. For the data analysis, use will be made of HDM-4. Members of the RAMS task force will be given training
in HDM-4 and will be involved in carrying out strategy and programme analyses of the collected data. It is
recommended that initial training in HDM-4 be provided by the ADB RAMS consultant with follow-up training
provided abroad through a certified HDM-4 training institute. The RAMS consultant should be made
responsible for carrying out an HDM-4 strategy and programme analysis and preparing 5-year rolling
maintenance investment plans for two consecutive years, involving staff of the RAMS unit throughout this
process as part of their on-the-job training.

49. For the future data collection, management and analysis, the RAMS unit will require an annual budget
allocation. This will cover the costs of the annual road network surveys, as well as costs of preparing maps and
reports and keeping the database (including the remote access) and data collection equipment running
properly. It is expected that the RAMS unit will require a minimum annual budget of $100,000 (excluding staff
costs), although this estimation will need to be refined by the ADB RAMS consultant. The costs of data
collection in the initial years will be higher as all data needs to be collected (in subsequent years, much of the
inventory data will remain the same and only needs to be checked).

5. PLANNING
50. The process of planning and budgeting follows former Soviet procedures. An inspection of the road
network is carried out by the GUSADs together with the traffic police each year. This identifies the maintenance
needs, which are forwarded to the Regional Offices that prepare a budget request using standardized unit
rates. The budget requests are forwarded to the Road Construction & Maintenance Department which
consolidates them and prepares a budget request for the entire road network. This budget request does not
include the full maintenance needs as identified by the GUSADs but includes a more limited subset (MOT does
not consider it appropriate to ask for too much funding). For 2019, for instance, the budget request was for
TJS 120 million, although the maintenance needs were mentioned to be around TJS 300 million. Once the MOF
has approved the annual budget allocation (generally in the order of 20% of the required amount – TJS 68
million for 2019), the actual workplan is prepared. The division of the approved budget between the different
GUSADs is on the basis of the size of the network they manage (by class) and the terrain they are located in
(traffic volumes and road conditions do not appear to be taken into consideration in this process). This division
amongst the GUSADs is done using Excel.

51. The introduction of the RAMS will introduce a major change in the planning process. Where currently the
planning and budgeting is done at the lowest levels and sent to the regional offices and MOT where it is
consolidated, in the RAMS only the data collection is done at the lowest levels and the planning and budgeting
is done at central level. Data on maintenance needs is collected for the entire network and is used as the basis
for determining the required budget (applying similar unit rates).

52. The RAMS allows the budget requirements for the entire network to be determined, but it may also be
used to determine the impact of different budget levels on future road network conditions. As such it can form
a powerful tool in negotiations with MOF regarding appropriate budget allocations, ensuring as a minimum
that road network conditions do not deteriorate due to a lack of maintenance funding (requiring more costly

13
interventions in the future). For the budget request to be provided to MOF, it is recommended that this include
a presentation of the full budget needs (unconstrained budget) as well as a couple of lower budget scenarios
that are more in line with the expected budget availability, indicating for each what the expected impact will
be on road conditions in the immediate and medium term.

53. Once the annual budget allocation has been approved by MOF, the RAMS may be used to determine how
this is best allocated to the different roads. This may be done on a purely economic basis using HDM-4, but it
may also be combined with other criteria defining the division of funds between different regions and GUSADs.
Care needs to be taken that the division of the available budget is not dictated too much by equality criteria
and the desire to provide funding to all GUSADs, as this is likely to reduce the overall benefits for road users as
funding is no longer allocated to the highest priority roads in the country. It may therefore be necessary to
step away from a system where budgets are allocated to GUSADs based on existing criteria, and move to a
system where most of the funding is allocated directly to specific roads according to economic criteria
(depending largely on traffic volumes and road conditions).

54. While the development of a RAMS is relatively simple, integrating the RAMS into the planning and
budgeting process and ensuring that budget allocations ultimately follow the priorities identified by the RAMS,
is much more complicated. These are important aspects of the RAMS development that are often overlooked.
The ADB RAMS consultant will need to carry out a detailed review of existing planning procedures, steps and
timeframes to see how the RAMS may be integrated into the process. This will include practical aspects such
as ensuring that data collection and analysis are carried out in time for the RAMS results to feed into the
planning process. But it will also include more complicated aspects such as changing the criteria based on
which approved budgets are divided and allocated to the different GUSADs. It will also identify any procedural
or legal issues that may prevent the results from the RAMS analysis from being integrated into the planning
process.

55. Apart from a detailed assessment of the current planning procedures, it is therefore recommended that
the ADB RAMS consultant be required to prepare a detailed Road Maintenance Planning & Budgeting Manual
that describes the different steps to be undertaken in the planning process, and how the RAMS fits into this.
The manual will need to identify the units responsible for providing the different data, and the timeframes for
doing so, as well as defining the criteria to be used for prioritization and allocation of available budgets. This
should not be seen as a purely technical document, as it largely depends on policy decisions to be taken by
MOT regarding priorities in funding allocations.

56. In support of the maintenance planning, it is recommended that the ADB RAMS consultant be required
to prepare a strategy analysis identifying the maintenance investment priorities for different road categories
based on important characteristics such as traffic volume, class, surface type and condition. This may be used
as the basis for the allocation of funding in the future. The consultant should also be required to prepare a
rolling 5-year maintenance investment plan identifying specific road sections to receive maintenance funding.
This should be required for at least two consecutive years (for fiscal years 2022 and 2023).

6. FINANCING
57. Funding for road maintenance is an issue in all countries, and Tajikistan is no exception. In Tajikistan the
level of maintenance funding is exceptionally low, however, and lack of adequate maintenance funding is by
far the most serious issue facing the sustainability of the road network. This section looks at the current level
of road maintenance funding and compares this to the estimated maintenance needs. It goes on to look at the
creation of a Road Fund as promoted by various development partners, and describes the different road user
charges and their potential for generating revenue for such a Road Fund.

MAINTENANCE BUDGET

58. MOT receives funding for road maintenance from the Republican Budget. This forms around 5-7% of the
total MOT budget. Most of the road maintenance funding is spent on international roads and to a lesser extent

14
on republican roads. The budget allocations for 2019 and previous years are indicated in Table 7. For 2019
MOT received a road maintenance budget of TJS 68 million against a budget request of TJS 120 million and an
estimated budget need of TJS 300 million. Although the amount in US dollars has remained more or less
constant over the past decade (decreasing slightly compared to 2010), the amount in Somoni has doubled in
that same period, increasing annually. This indicates that the fall of the exchange rate of the Tajiki Somoni has
resulted in significantly lower purchasing power of the allocated maintenance budget.

Table 7 MOT road maintenance budgets (million)

Item 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
Road maintenance budget (TJS) 34.0 39.0 46.8 50.6 57.2 59.6 62.9 60.2 60.8 68.0
Total MOT budget ($) $145.6 $196.9 $196.4 $216.8 $263.6 $154.0 $109.9
Road maintenance funding share
5.4 8.7 5.0 4.9 4.4 6.2 7.0
of total MOT budget (%)
Source: ADB TA-8945 TAJ: Road Sector Financing Report

59. According to the World Bank, the GDP of Tajikistan for 2017 is $7.146 billion. This means that the current
road maintenance budget of TJS 68 million ($7.2 million) forms approximately 0.1% of the GDP. This is very
low compared to other countries.

MAINTENANCE NEEDS

60. The estimation of the required maintenance funding varies wildly. Other countries in the region
reportedly spend 0.4% of GDP on road maintenance, which would imply a maintenance allocation in Tajikistan
of approximately $30 million per year.

Table 8 Road maintenance funding as percentage of GDP

Country GDP Nominal Road Network Maintenance Budget Maintenance budget
($ billion) (km) ($ million) (% of GDP)
Armenia $10.6 7,704 $44.0 0.42%
Kyrgyzstan $6.6 18,585 $29.3 0.45%
Tajikistan $7.9 14,146 $7.7 0.09%
Georgia $14.0 6,824 $60.0 0.43%
Source: ADB TA-8945 TAJ – RAMS Action Plan

61. MOT itself estimates that it needs around TJS 300 million (approximately $40 million) to address all
maintenance needs in the road network (this should be compared to the actual budget request of TJS 120
million). This is in line with what is mentioned by the GUSADs which state that they receive approximately 20-
30% of what they actually need.

62. A TRACECA study (LOGMOS Master Plan – Road Sector Overview) carried out in 2014 estimates that the
required minimum annual expenditure on road maintenance in Tajikistan is about $47 million, and that an
additional annual expenditure of $135 million per year would be needed to progressively remedy the backlog
of rehabilitation.

63. The Transport Sector Master Plan (2015) mentions that about 1.8% of Tajikistan’s GDP is required for
roads, with 1.1% for rehabilitation and at least 0.7% for periodic and routine maintenance. This would imply a
maintenance funding requirement of $50 million per year based on the 2017 GDP of $7.146 billion. An HDM-
4 analysis carried out as part of the RAMS Action Plan under TA-8945 TAJ estimated that a budget allocation
of $32 million per year (0.4% of GDP) would allow the condition of the road network to be stabilized (with
higher level roads improving and lower level roads deteriorating), while a budget of $55 million per year (0.7%
of GDP) would allow a gradual improvement of the entire road network.

15
 Figure 7 Expected future road network condition under different budget scenarios

$7.7 million (0.1% GDP) $32 million (0.4% GDP) $55 million (0.7% GDP)
100% 100% 100%
90% 90% 90%
80% 80% 80% 34%

70% 57% 70% 57% 58% 70% 57%

60% 60% 60%
85%
50% 50% 50% 34%
40% 40% 40%
30% 25% 30% 25% 25% 30% 25%
20% 20% 20%
32%
10% 17% 8% 10% 17% 17% 10% 17%
0% 6% 0% 0%
2016 2036 2016 2036 2016 2036

Good Fair Poor Good Fair Poor Good Fair Poor

Source: ADB TA-8945 TAJ RAMS Action Plan

64. As the road network is gradually rehabilitated and more roads become maintainable, the maintenance
costs are expected to increase. A basic estimation of the maintenance needs for the entire road network
managed by MOT based on average unit costs results in a total long-term annual maintenance cost in the order
of $110 million. This includes routine and winter maintenance for all roads each year, as well as periodic
maintenance (regravelling or overlays) every 5-15 years (depending on the road class)8. This is a very rough
estimate but indicates the order of magnitude of the long-term maintenance needs. What is noteworthy here
is that the local roads make up half the total maintenance costs, with international and republican roads
requiring an annual maintenance budget of approximately $60 million per year in the long term.

Table 9 Estimated long-term maintenance needs

Maintenance Unit cost ($/km/year) Length (km) Cost ($ million/year)

type Inter- Repub- Local Inter- Repub- Local Inter- Repub- Local Total
national lican national lican national lican
Routine BT $1,500 $1,500 $1,500 2,993 1,666 5,672 $4.5 $2.5 $8.5 $15.5
Routine GR/ER $1,000 $1,000 $1,000 355 463 3,071 $0.4 $0.5 $3.1 $3.9
Winter $500 $500 $500 3,348 2,129 8,744 $1.7 $1.1 $4.4 $7.1
Periodic BT $10,000 $7,500 $5,000 2,993 1,666 5,672 $29.9 $12.5 $28.4 $70.8
Periodic GR $5,000 $3,500 $2,500 355 463 3,071 $1.8 $1.6 $7.7 $11.1
Tunnel $200,000 17.9 $3.6 $3.6
Bridge $30,000 42.7 $1.3 $1.3
Total $43.1 $18.1 $52.0 $113.2
Source: Consultant’s processing of data

65. More accurate estimations will be possible once the RAMS is in place and more detailed data is available
for the road network. It is recommended that the ADB RAMS consultant be made responsible for preparing an
initial assessment of the total maintenance needs based on the first run of HDM-4 using the data for the 1,000
km of roads or more to be collected. This will need to be expanded on later as data is collected for a greater
portion for the road network and the HDM-4 is further calibrated.

8Overlays every 7 years for international roads, every 10 years for republican roads and every 15 years for local roads. Regravelling
every 5 years for international roads, every 7.5 years for republican roads and every 10 years for local roads.

16
ROAD FUND

66. What is clear from the different estimations listed above is that the current amounts of road maintenance
funding ($7.2 million) are far below what is needed ($30-$110 million), forming only a quarter of the lowest
estimated requirement. There is a clear need for additional road maintenance funding. This is not easily
achieved, however. Current funding comes from the Republican Budget, where it has to compete for financing
with many other sectors. Development partners have therefore been promoting the idea of introducing a Road
Fund financed from hypothecated or earmarked road user charges9, ensuring a dedicated funding stream that
is not or hardly affected by other needs.

67. The introduction of dedicated road user charges also introduces a “user pays” principle. This assumes that
road users are willing to pay for the use of the road, if they see that this results in better road conditions and
related benefits for them in terms of reduced road user costs. This concept also assumes that road users should
pay according to the amount of damage they cause to the road and the amount of use they make of the road.

68. A Road Fund existed in Tajikistan from 1992-1999 but it was terminated to meet the International
Monetary Fund (IMF) requirements to maintain a single consolidated State Budget. All taxes and fees from
road users currently go to the Republican Budget and are under the jurisdiction of the Ministry of Finance, with
no earmarking of funds for road maintenance financing.

69. ADB and EBRD have been promoting the idea of reintroducing a Road Fund, this time as an autonomous
second-generation fund set up as a state-owned company and run by a Board of Directors. It is not clear what
the position of the government is in this regard. MOF appears to be reluctant to introduce any rigid
hypothecation of revenues, and also sees little room for the reallocation of existing revenues away from other
needs. They instead recommend that MOT provide a better justification for increased road maintenance
funding, which may then be approved if the financial situation allows for it.

70. It is recommended to use the RAMS to prepare a detailed justification for the budget request, presenting
the expected impact of different budget scenarios on longer-term road network conditions and rehabilitation
needs. This has proven to be a strong tool in negotiating increased road maintenance funding with the finance
ministries in other countries. However, with the limited funding available in the Republican Budget, even such
improved justification is unlikely to result in the budget increases required to ensure proper maintenance of
the road network. Additional sources of funding will be required.

71. The most common source of funding for road maintenance are road user charges. Especially where a
Road Fund exists, road user charges are commonly used to finance it. Road user charges are often introduced
as hypothecated revenue for a Road Fund, but they may also simply be introduced as earmarked revenue for
the Republican Budget which is (mostly) allocated to road maintenance in response to the budget request and
related justification.

72. Pakistan is an example of the latter, where a Road Maintenance Account (RMA) was introduced rather
than a Road Maintenance Fund. The RMA is financed from road user charges, general budget allocations and
development partner loans. The NHA has a RAMS and uses HDM-4 to prepare annual maintenance cost
estimates. Budget requests are accompanied by a justification showing the expected impact on future road
network conditions under different budget scenarios. The annual allocation to the RMA continues to be
determined by the Ministry of Finance as part of the annual budget that is approved by parliament, and largely
corresponds to the requested budget.

9 In the case of hypothecated road user charges, the collected revenue goes directly to the Road Fund without passing through the
Republican Budget. These are extrabudgetary funds. In the case of earmarked road user charges, the legislation identifies revenue from
certain road user charges as being intended for financing the Road Fund. However, the revenue first goes to the Republican Budget
from where it is allocated to the Road Fund as part of the annual budget. Although the revenue from these road user charges is
earmarked, the government and Parliament have the authority to adjust the actual allocation to the Road Fund if circumstances require
this.

17
73. Road user charges are not only promoted by development partners. Resolution No. 165 of April 2011 on
the National Program for the Development of the Transportation Sector of the Republic of Tajikistan up to 2025
states that maintenance of roads will be provided entirely out of the contributions from the local road users.
It foresees gradually increasing financing adding up to $63.5 million in the period up to 2015 (average of $16
million per year), $166.8 million up to 2020 (average of $20 million per year), and $287.4 million up to 2025
(average of $24 million per year).

74. The following sections deal with some of the possible road user charges that exist or may be introduced
in Tajikistan, identifying the potential revenue that may be generated. This includes access charges that have
to be paid by road users to gain access to the road network (irrespective of the amount of usage) and usage
charges that are paid by road users according to the degree of usage. An example of the former is an annual
vehicle registration tax that has to be paid by each vehicle owner irrespective of the usage of the vehicle. An
example of the latter is a toll that has to be paid each time a road user makes use of the road. The Transport
Sector Master Plan (2015) estimates that the resources generated from road users - including all tariffs, fees,
charges, and excise taxes - amount to TJS 73 million per year. However, the actual revenue from existing road
user charges is significantly higher as is shown below.

ROAD USER TAX

75. The Road User Tax in Tajikistan is applied to corporate and individual taxpayers with a gross revenue
exceeding TJS 600,000, who are required to pay a tax over the costs of their goods and services (with a
minimum of 70% of their gross revenue). Retail tax and VAT are excluded from the tax base, and state bodies,
religious and cultural organizations are exempted. The tax rate is 0.25% for trading, procurement and supply-
sale activities, and 1% for other activities (down from 0.5% and 2% in 2014). The tax is administered in a similar
manner to the VAT. There has been talk of abolishing the Road User Tax for some time, but this keeps getting
postponed and the tax is now expected to remain at least until the end of 2021.

76. Originally the Road User Tax provided financing for the Road Fund, but with the abolishment of the Road
Fund in 1999, the revenue reverted to the general Republican Budget. According to the “Report on the
Implementation of the Republican Budget in 2018”, the revenue from the Road User Tax reached TJS 387.7
million (approximately $42 million).

77. Although this tax was introduced to finance the road sector through the Road Fund, it is strictly speaking
not a road user charge and is instead a business tax. It is neither an access nor a usage charge, as the payment
is not directly related to the access to or usage of the road network. Although the revenue from the Road User
Tax could cover a significant portion of the required road maintenance costs, the future of this revenue source
is uncertain. The Ministry of Finance also appear to be reluctant to introduce any earmarking of this or any
other existing revenue source for road maintenance.

TOLLING

78. Tolling is a common source of road user charges whereby the road users are directly charged for the use
of the road (usage charge). Tolling is generally applied only in roads with higher traffic volumes, as the
collection costs would otherwise be too high in relation to the revenue. Even in high volume roads the
collection costs can form up to 20% of revenue.

79. Tolling is often applied only in roads for which alternative non-tolled routes exist. In such cases, toll rates
may be higher that the costs of operating and maintaining the road and may also serve to pay back the
financing of the road or to cover the costs of maintaining other roads. Where no alternative routes exist, tolling
may still be applied but is then often limited to covering the costs of operating and maintaining the road
concerned.

80. There is currently one toll road in Tajikistan that is operated as a concession contract on the 368 km main
road north from Dushanbe through Khujand to the border post at Chanak, from where it connects to Tashkent
in Uzbekistan. The highway was originally upgraded with a Chinese loan of $280 million. The concession

18
contract started on 1 April 2010 and covers 337 km of the road10. It is operated by Innovative Road Solutions
LTD (IRS) with a contract period of 30 years. It is not clear what the scope of responsibility of the concessionaire
is, but it appears to include routine/winter and periodic maintenance as well as some rehabilitation and
upgrading (it appears that the maintenance of the two large tunnels is not included).

81. The toll road has 6 tolling stations along the road. Differentiated toll rates are applied for four vehicle
categories, although it is not clear what vehicle types are included in each category. Toll rates range from an
average of TJS 0.2/km for category I ($0.02/km) to TJS 1.9/km for category IV ($0.20/km). Local residents
receive a considerable discount. The total toll cost for the entire route varies from TJS 59 for category I
(approximately $6) to TJS 638 for category IV (approximately $70).

Table 10 Toll rates on Dushanbe-Khujand-Chanak toll road (TJS)

Vehicle type Category I Category II Category III Category IV

Varzob 3 12 30 38
Hushory 21 59 147 185
Shahristan 23 69 173 217
Istaravshan 6 21 53 66
Dehmoy 2 6 30 38
Poul Chorouh 4 21 60 94
Total 59 188 493 638
Source: www.irs.tj June 2019
Figure 8 Toll stations in Varzob and Istaravshan

82. The road is managed by three regional offices of IRS in Varzob, Zarafshon and Khujand. These are each
responsible for specific sections of the road. The section chainages and their lengths are indicated in the table
below.

Table 11 IRS regional offices and related road sections

IRS Regional Office Start End Length (km) From - To

Varzob 10+000 73+300 63.3 Varzob - Anzob/Isiqlol tunnel
78+730 119+000 40.3 Anzob/Isiqlol tunnel - Ayni
Zarafshon
119+000 160+437 41.4 Ayni - Sharistan tunnel
165+726 232+000 66.3 Sharistan tunnel - Istaravshan
Khujand 232+000 297+600 65.6 Istaravshan - Khujand
308+200 368+200 60.0 Khujand - Chanak
Source: www.irs.tj June 2019

10 Excluded sections involve the road sections through Dushanbe and Khujand as well as the two large tunnels.

19
83. Information on the revenue collected is not readily available. Traffic data published by MOT for
Anzob/Isiqlol tunnel show an average daily traffic volume of 2,000 vehicles per day11. It is assumed that this
traffic volume consists of 60% category I, 20% category II 10% category III and 10% category IV vehicles. It is
further assumed that the reported 2,000 vpd involves vehicles travelling the full length of the toll road, and
that 25% of the road length around Dushanbe and Khudjand will have a significantly higher traffic volume of
4,000 vpd. Based on these assumptions, annual revenue will be in the order of $21 million per year,
corresponding to $64,000/km/year. This easily covers routine and winter maintenance as well as periodic
maintenance, leaving sufficient funding for rehabilitation and upgrading works.

Table 12 Estimated annual toll revenue Dushanbe-Khujand-Chanak

Category I Category II Category III Category IV Total

Traffic composition 60% 20% 10% 10% 100%
Average traffic (100% length) 1,200 400 200 200 2,000
Extra traffic (25% length) 2,400 800 400 400 4,000
Toll rate total length 59 188 493 638
Daily revenue 106,200 112,800 147,900 191,400 558,300
Annual revenue (TJS) 38,763,000 41,172,000 53,983,500 69,861,000 203,779,500
Annual revenue (USD) $4,080,316 $4,333,895 $5,682,474 $7,353,789 $21,450,474
Source: Consultant’s processing of data

84. The ADB foresees the introduction of new toll roads under its upcoming CAREC Corridors 2, 3, and 5
(Obigarm–Nurobod) Road Project. The project aims to identify two road sections where tolling will be piloted.
These toll roads will not be concession contracts but will be traditional toll roads where tolls are collected by
MOT (either directly or outsourced) and used to finance the maintenance of the road. Toll rates may be set at
a level where they are adequate to cover only the maintenance costs of the toll road, or increased to also cover
maintenance costs for other roads.

85. The challenge that will be faced in piloting this approach to tolling, will be to ensure the toll revenue is
allocated to road maintenance. Currently any revenue collected by the government will flow into the
Republican Budget, and this will also be the case for the toll revenue. It is possible to earmark the toll revenue
for road maintenance by means of the legislation creating the Road Fund, but this does not fully guarantee
that the toll revenue will be fully allocated to road maintenance. Toll revenue also allows for hypothecation,
with the Road Fund for instance directly collecting the toll revenue and depositing this it in the bank account
of the Road Fund as extra-budgetary revenue. However, it is unlikely that this will be considered acceptable to
MOF.

86. Potential revenue will depend on the road sections to be selected for the tolling (length and traffic
volume) and the toll rates to be applied. It is likely that the tolling will be piloted in international roads,
specifically the CAREC corridors that carry the most traffic. Toll rates may be set at similar levels as for the toll
concession, or they may be adjusted to fit the expected maintenance costs for the tolled road sections (and
possibly other roads in the area).

Table 13 CAREC corridors in Tajikistan

CAREC Corridor Section Remarks

2 and 5 Dushanbe east to Karamyk and Kyrgyz Largely completed, lower traffic
2, 5 and 6 Dushanbe south to Kurgonteppa and Afghanistan Under construction, important border crossing
3 Dushanbe west to Tursonzoda and Uzbekistan Completed, important border crossing
6 Ayni east to Panjakent and Uzbekistan Completed, important border crossing
6 Dushanbe north to Khujand and Uzbekistan Existing toll concession

11 Reported traffic volumes for 2019 are significantly lower than for 2018, although it is not clear what the reason for this is.

20
FUEL EXCISE TAX

87. A fuel excise tax is a very common source of funding for road maintenance. It is a road user charge that
people pay according to the usage of the roads (usage of their vehicles). It is applied as an earmarked or even
hypothecated funding source for most road funds. It is so commonly used because it is relatively easy and
inexpensive to collect, and because it increases with increased vehicle ownership and fuel consumption. Thus,
as the damage from vehicles increases, so does the revenue for road maintenance funding.

88. China is a prime example, introducing a fuel tax in 2009 that replaced existing tolls and vehicle registration
fees that were previously used to finance road maintenance. In the Netherlands the fuel excise tax is mainly
used to finance the general state budget. However, in 1995 the government increased the fuel excise tax by
EUR 0.11/litre12 ($0.13/litre) specifically to finance road development and maintenance. Currently the total
fuel excise tax in the Netherlands amounts to EUR 0.77/litre ($0.90/litre) for petrol and EUR 0.48/litre for diesel
($0.56/litre), only part of which is allocated to the road sector. In both cases the revenue goes to the national
budget, to be subsequently allocated to road maintenance. Neither country has a road fund.

89. Tajikistan already collects a fuel excise tax. According to Resolution 102 of February 15th 2014 “on the
Approval of the Rates of Excise Duty on Certain Goods, Produced in the Republic of Tajikistan and Imported to
the Territory of the Republic of Tajikistan”, the excise tax for petrol is 55 euros per tonne (just over $0.04/litre),
while for diesel it is much lower at 8 euros per tonne (just under $0.01/litre). The volumes of petrol and diesel
imported and produced domestically in 2015 are indicated in the following table. Based on these fuel
consumption volumes, the revenue from the fuel excise tax is estimated to be approximately $15 million.

Table 14 Fuel usage in Tajikistan (2015)

Fuel Metric tonnes Litres Excise tax

Petrol 215,731 292,715,468 $12,933,091
Diesel 215,636 258,909,185 $1,880,346
Total 431,367 551,624,653 $14,813,437
Source: TA-8945 TAJ Road Sector Financing Report

90. However, not all fuel consumption is for road use. If it is assumed that 100% of the petrol and 60% of the
diesel is used by road vehicles, the automotive fuel consumption would be approximately 345,000 tonnes or
450 million litres. The resulting revenue from the current fuel excise tax for automotive use would still be in
the order of $14 million.

91. It is unlikely that the existing fuel excise tax revenue will be allocated to road maintenance, and it may
therefore be required to introduce an additional fuel excise tax specifically earmarked for road maintenance.
This would be a similar approach as applied in the Netherlands in 1995 and China in 2009. An increase to the
fuel excise tax of $0.05/litre for petrol and $0.03/litre for diesel would result in an additional revenue of
approximately $19 million.

92. The reallocation of the existing fuel excise tax or the allocation of an increase to the fuel excise tax to road
maintenance should be considered in Tajikistan. Especially if a Road Fund is to be introduced, a fuel tax is a
straightforward source of funding that has proven to work in many countries. The increased excise tax may be
introduced gradually over a couple of years, avoiding large jumps in the pump prices. It may also be timed to
coincide with decreasing fuel prices, to further minimize the jumps in pump prices.

VEHICLE TAX

93. Annual vehicle registration fees are a common source of revenue that can be used to finance road
maintenance. Vehicle ownership has increased significantly in Tajikistan, from 3.3 vehicles per 100 people in
2008 to 5.5 in 2015. The total number of vehicles has increased from 185,684 in the year 2000 to 434,652

12 Then equivalent to NLG 0.25/litre and referred to as “the quarter of Kok” (the finance minister at the time).

21
vehicles in 2015 (an average annual growth rate of 5.8%). Passenger cars make up roughly 85% of the vehicles,
with trucks forming 10%, buses 4% and motorcycles the remaining 1%.

Table 15 Number of registered vehicles (December 2015)

Light Truck Bus Special Motorcycle Trailer Semi-trailer Total

vehicle vehicle
370,711 39,160 16,186 4,093 2,926 1,423 153 434,652
85% 9% 4% 1% 1% 0% 0% 100%
Source: MOT 2016

94. In Tajikistan the Vehicle Tax is applied according to the engine capacity in horsepower, which is multiplied
by the minimum wage (currently TJS 40) over which a variable rate is applied according to the type of vehicle
as indicated in the table below (annual vehicle tax payment = minimum wage * tax rate * horsepower). The
Vehicle Tax is collected at the time of the annual technical inspection of the vehicle.

Table 16 Vehicle Tax rates by vehicle type

Vehicle type Tax rate

Motorcycles and scooters 2.5%
Vans and pick-ups on the basis of light motor vehicles 6.0%
Light motor vehicles 7.5%
Buses (up to 12 seats) 7.5%
Buses (for 13-30 seats) 8.5%
Buses (over 30 seats) 9.5%
Trucks and other transport vehicles with weight-carrying capacity up to 10 tons 11.0%
Trucks (with weight-carrying capacity from 10 up to 20 tons) 12.5%
Trucks (with weight-carrying capacity from 20 up to 40 tons) 13.5%
Trucks (with weight-carrying capacity from 20 to 40 tons) 14.5%
Tractors, vehicles with engines for construction, except for use in agriculture 2.0%
Source: Tajikistan Tax Code

95. According to the “Report on the Implementation of the Republican Budget in 2018”, the revenue collected
from the Vehicle Tax for 2018 amounted to TJS 189.2 million (approximately $20 million). The Vehicle Tax in
Tajikistan is collected at local level and the revenue is allocated to the local budgets. Although there are
countries where vehicle taxes are collected locally and subsequently allocated to the a central government
entity for the maintenance of the road network, this involves deconcentrated local offices under a central
ministry responsible for the collection of the fees. Where the collection has been decentralized to local
authorities instead of deconcentrated to local offices under a central ministry, and especially where the
collected revenue goes to the local budgets, such allocation to a central ministry does not happen. The revenue
is considered local revenue, with allocation to be decided by the local authorities. As such, the Vehicle Tax is
not considered a suitable source of funding for the maintenance of international and republican roads or for
financing a Road Fund. However, it does reflect a considerable financial capacity of the local governments to
finance the maintenance of the local roads.

OVERLOADING FINES AND FEES

96. Overloading fines and fees form a very suitable road user charge, as it is applied to those road users
causing most of the damage to the road network. Overloading fines do exist in Tajikistan, but with only a
limited number of weigh scales, enforcement is limited.

97. The EBRD consultants for the Review of Road Standards and Road Sector Institutional Reform visited a
few of these weigh stations. The weigh station at Hissar was found to only check semi-trailers, despite the fact
that single axle trucks tend to be the ones causing the most damage to the road due to the high axle loads.
International trucks were not checked as these were assumed to have been checked at the border. Axle loads
were measured using portable weigh scales, but only gross weight was controlled (sum of the axle loads).
Drivers of overloaded trucks (exceeding 40 tonnes) had to pay a fine before they could continue their journey.

22
At the Kyrgyz border there is a new weigh station, but this can only check the gross weight, not the individual
axle loads.

Figure 9 Hissar and Kyrgyz border weigh stations

Source: EBRD Review of Road Standards and Road Sector Institutional Reform – Inception Report (2016)

98. The toll concessionaire also operates a low speed dynamic weigh station at Warzob, 39 km from
Dushanbe. This includes both axle load and gross weight control, comparing these to Tajik standards. Drivers
of overloaded vehicles are required to pay the fines directly before being allowed to continue their journey.

Table 17 Weight limits and applied fines

Axle type Axle weight Fine

Single axle 10.0 Tonnes TJS 35.0
Double axle 18.0 Tonnes TJS 49.0
Triple axle 22.5 Tonnes TJS 70.0
Gross weight 40.0 Tonnes TJS 52.5
Source: EBRD Review of Road Standards and Road Sector Institutional Reform – Inception Report (2016)

Figure 10 Warzob weighing station

Source: EBRD Review of Road Standards and Road Sector Institutional Reform – Inception Report (2016)

99. It is not clear what happens with the revenue collected from the overloading fines. For the fines applied
by the traffic police, it would appear that these go to the Republican Budget. For the toll concession road, it
would appear that the fines go to the concessionaire. Although overloading fines are considered an
appropriate revenue source for road maintenance, its use is not considered a very practical one. The fines are
collected by the traffic police (or the concessionaire) and as such the collection costs need to be covered from
those fines as well. Currently the levels of the fines are very low at $6.7 for exceeding the gross weight, and

23
revenue after covering collection costs will likely be very low. It is considered that the fines should be set
significantly higher, better reflecting the amount of damage caused by overloading. It is recommended that
the RAMS and HDM-4 analysis be used to identify a suitable level of overloading fines based on axle loads and
gross weights, which are in line with the repair costs of the damage caused by such overloading.

VEHICLE IMPORT DUTIES

100. Import duties and excise taxes on vehicles may form another source of funding for road maintenance.
This is considered an access charge, as it has to be paid by road users in order to gain access to the road
network, irrespective of the degree of usage. This may include customs duties as well as initial registration
charges for vehicles and excise taxes relating to the initial purchase of a vehicle.

101. In Tajikistan the customs duty on vehicles ranges from 5% to 12% of the customs value as shown in the
table below. Tajikistan also knows an excise tax for passenger vehicles (commodity code 8703), although no
rate is currently applied. Information on the total revenue from the customs duties and initial vehicle
registration fees has not yet been obtained. It is recommended that further information on this revenue and
the potential revenue from increases to these rates be determined as a next step.

Table 18 Customs duties

Vehicle type Passenger vehicles Passenger vehicles Cargo Construction Dump

& cargo vans (>10 passengers) vehicles vehicles trucks
Customs duty 10% 12% 8% 8% 5%
Source: Resolution 399 of 8 August 2018 on the rates of import customs duties in the Republic of Tajikistan

7. IMPLEMENTATION
102. In 2016, there were 329 registered contractors in Tajikistan licensed to undertake road construction and
maintenance works. Amongst these, 18 are international contractors, 154 are local private contracting
companies and 141 are state-owned companies (including 79 companies assumed to be involved mainly in
construction, and the 62 GUSADs involved in maintenance).

Category I Category II Category III Category IV Total

International companies 13 4 1 18
Limited Liability Companies (LLC) 27 41 75 6 149
Joint Stock Companies (JSC) 1 3 1 5
Others 1 6 7 2 16
State-owned 10 8 21 40 79
GUSADs 62 62
Total 52 62 105 110 329
Source: EBRD Review of Road Standards and Road Sector Institutional Reform - GUSADs’ Transformation Report March 2017

The contractors are registered by category. Category I contractors are able to carry out large bridge
construction works (>60m) as well as construction works on category I and II roads and can carry out works
supervision. Category II contractors are allowed to work on category I-IV roads in mountainous areas and can
carry out surveys. Category III contractors can carry out works that are not restricted to the higher contractor
categories and can carry out quality control and road performance surveys. Category IV contractors can only
carry out off-carriageway works and road works for lengths of less than 1 km, as well as visual assessments. It
is interesting to note that the GUSADs are all registered as category IV contractors.

The RAMS analysis is likely to result in a much higher priority being given to periodic maintenance (rather than
only routine maintenance and rehabilitation). This will require the creation of a domestic capacity to carry out
such periodic maintenance, both in terms of experience and skilled staff, as well as equipment. In Pakistan the
introduction of the RAMS resulted in a shift of funding allocation towards periodic maintenance, which formed
59% of the annual maintenance budget in 2014-15. As a result of the increased periodic maintenance, the
average road network roughness has reduced from an IRI of 8.0 in 2000 to 5.6 in 2014. However, it took several

24
years for contractors to gain experience and skills and to acquire the necessary equipment to carry out the
large volumes of periodic maintenance. This will also be a challenge in Tajikistan, and a strategy needs to be
developed to create that capacity. A decision will also need to be taken whether this capacity should be created
in the GUSADs or with contractors.

GUSADS

103. Road maintenance in Tajikistan is mostly carried out through force account by the GUSADs under MOT.
The GUSADs are set up as State Institutes under MOT and have staff and basic equipment to carry out routine
and winter maintenance and carry out basic repairs. Periodic maintenance (surface treatments, overlays) is
not carried out much. Emergency maintenance is also often carried out by the GUSADs, ensuring that roads
remain open.

104. The 62 GUSADs each cover an average road network of 230 km, ranging from 60 km to over 600 km
(Murghob district in GBAO that encompasses over a quarter of Tajikistan’s territory). They have an average of
just over 40 staff, ranging from 20 to 100. The total staff adds up to just under 2,700, approximately half of
who are semi-skilled workers. Staff tend to be aged around 50 years old. The equipment of the GUSADs is to a
large extent old and unserviceable, although some GUSADs have recently received equipment from JICA and
EBRD.

Table 19 Number of GUSADs and road length (km) by Regional Office

Gissar Kulyab Kurgonteppa Sughd Korugh Rasht

GUSADs (#) 9 10 13 14 9 7
International roads 287 289 435 732 1,297 299
Republican roads 411 432 325 414 384 175
Local roads 1,213 1,670 1,980 2,306 1,031 466
Total roads 1,911 2,391 2,740 3,452 2,712 940
Source: MOT 2016

105. Due to the fact that the GUSADs have large numbers of staff and old equipment, their efficiency is not
very high. Approximately 40% of the road maintenance budget is spent on salaries and social protection fund
payments for the staff of the GUSADs, with just over half the road maintenance budget spent on actual
maintenance works (including centrally executed funds for the Districts of Republican Subordination). The staff
costs remain more or less the same irrespective of the maintenance budget, which is a major weakness of the
approach of using GUSADs (outsourced contracts will only fund a portion of the staff costs of contractors, with
the rest financed from other contracts – where less staff is required, the contractor will reduce the staff
numbers). With low road maintenance budgets, this significantly reduces the amount of funding actually spent
on road maintenance.

Table 20 Road maintenance budget allocation

Budget item 2013 2014 2015 2016

Republican road maintenance 10,767,161 10,769,356 12,189,904 14,774,401
Local road maintenance 4,653,242 4,654,183 5,224,246 6,331,889
Major repairs 2,350,000 4,025,000 3,655,590 3,620,000
Salaries 15,403,459 18,321,384 18,321,384 20,677,504
Social Protection Fund 3,850,866 4,580,346 4,572,480 5,168,877
Centralized funds 10,887,000 11,853,234 12,615,692 9,400,000
Other costs 2,754,272 3,050,822 3,079,709 2,973,510
Total 50,666,000 57,254,325 59,659,005 62,946,181
Source: ADB TA-8945 TAJ: Road Sector Financing Report

106. The GUSADs are furthermore limited to carrying out routine and winter maintenance. This is reflected in
their registration as category IV contractors. Although some GUSADs have recently received surface treatment
equipment and pavers from JICA, this does not appear to be used currently. With the limited experience of
the GUSADs in carrying out more comprehensive maintenance works, careful consideration should be given to

25
whether their capacity should be increased to allow them to carry out periodic pavement maintenance, or
whether such works should be outsourced to contractors.

PERFORMANCE-BASED MAINTENANCE (PBM)

107. Although there is private sector participation in the road sector, this is largely restricted to road
construction works. Participation of private sector contractors in road maintenance is largely limited to the
four performance-based maintenance (PBM) contracts supported by ADB in two separate pilots. The contracts
are for 3 years (in the second pilot this included the option to extend for a further 3 years) and include routine
and winter maintenance. Emergency maintenance is included as a provisional sum. The PBM contracts are
separate from the contracts under which the roads were rehabilitated.

108. The first set of pilots were from May 2013 to June 2016. The contracts included the road section from
Nuromod to Nimich (Lot 1 – 120+000-193+000, 73 km) and from Vahdat to Obi Garm (Lot 2 – 0+000-76+000,
76 km). These were road sections that had been rehabilitated between 2005 and 2007 and were already
partially deteriorated. The contracts therefore also included some initial repairs. Lot 1 was awarded to
Innovative Road Solutions (IRS) and Lot 2 was awarded to Gayur. The average cost came to just under TJS
42,000/km/year ($7,600/km/year). Excluding the initial repair costs, the average cost still amounted to TJS
32,000/km/year ($5,800/km/year). This is considered quite high, but may be the result of the poor initial road
conditions and the unfamiliarity with the PBM approach.

Table 21 Costs of first phase PBM pilots (2013-2016)

Lot 1 (73 km) Lot 2 (76 km)

Contract Contract USD/km/year Contract Contract USD/km/year
price TJS price USD price TJS price USD
Routine 3,581,412 $648,995 $2,963 Routine 3,447,360 $624,703 $2,740
Winter 1,759,491 $318,841 $1,456 Winter 2,355,696 $426,881 $1,872
Repair 2,925,602 $530,155 $2,421 Repair 1,533,416 $277,874 $1,219
Emergency 1,156,400 $209,554 $957 Emergency 1,975,580 $357,999 $1,570
Total 9,422,905 $1,707,545 $7,797 Total 9,312,052 $1,687,457 $7,401
Source: PBM tender and contract documents

109. The second set of PBM pilots started in January 2018 and will run until December 2020. The contracts
include the road section from Sayron to Karamyk (Lot 1 – 217+670-306+630, 89 km) and from Vose to Khovaling
(Lot 2 – 0+000-86+800, 87 km). The road sections have been recently rehabilitated and the contracts therefore
do not include any repair works. Lot 1 was awarded to LLC “Dangarasokhtumon-10” and Lot 2 was awarded to
LLC “Shohroh – 2016”. The average cost came to just over TJS 13,000/km/year ($1,500/km/year). As such the
cost was much lower than the first pilots.

Table 22 Costs of second phase PBM pilots (2018-2020)

Lot 1 (89 km) Lot 2 (87 km)

Contract Contract USD/km/year Contract Contract USD/km/year
price TJS price USD price TJS price USD
Routine 1,500,785 $170,544 $639 Routine 1,899,209 $215,819 $809
Winter 1,533,984 $174,316 $653 Winter 850,290 $96,624 $362
Emergency 724,152 $82,290 $308 Emergency 400,631 $45,526 $171
Total 3,758,921 $427,150 $1,601 Total 3,150,130 $357,969 $1,341
Source: PBM tender and contract documents

110. Routine maintenance is paid on a performance basis according to compliance with a set of performance
standards with fixed monthly lumpsum payments. Payment deductions are made in case of non-compliance
with the performance standards. Winter maintenance is paid as a unit rate per kilometre with separate rates
defined for snow removal from the carriageway, applying salt/grit mixtures to the carriageway, and removing
snow from the shoulder and drainage system. Winter maintenance requires a work order issued by the Project
Manager. Emergency maintenance, repair works and periodic maintenance are also paid against unit rates.

26
Emergency maintenance also requires a work order to be issued, while the scope of work for repairs and
periodic maintenance (if any) is defined in the contract.

111. In the case of routine maintenance, the contractor is given a response time to correct any non-compliance
with the performance standards encountered during the inspection, after which the deducted payment is
included in the next monthly invoice13. This results in the need for additional inspections to check whether
defects have been corrected. More importantly, however, it means that contractors tend to do nothing until
the inspection identifies defects that need to be corrected, thus avoiding doing any activities that are not
required to avoid a deduction. This undermines the concept of performance-based maintenance, where the
contractor is required to take over some of the management responsibility for the road. It is strongly
recommended to remove the response times from any future PBM contracts where this is feasible. The
number and size of potholes, for instance, can easily be predicted on a monthly basis and the contractor should
correct such defects before they exceed the performance standard. Non-compliance during inspection would
then immediately result in a penalty.

112. The performance standards used in the PBM pilots are not considered suitable in all cases. To a certain
extent this is the result of the way the performance standards are formulated. For instance, the performance
standards allow up to 5 potholes larger than 10 cm in any dimension in a one-kilometre section of road, with
a maximum size of any pothole of 0.5m2. This means that there can be up to 5 potholes of up to 80 cm in
diameter in a one-kilometre section. For the ongoing contracts this implies more than 400 such enormous
potholes would be allowable in the road section under contract. This is excessive for a PBM contract. It is
recommended to revisit the performance standards and to ensure they reflect the service level that is required
for the road concerned. This will also require reformulating the performance standards to make them easier
to understand and assess.

113. The resulting penalties in case of non-compliance with the performance standards are set very low. In the
case of the number or size of the potholes exceeding the performance standard in a one-kilometre section, for
instance, the deduction is only 5% of the payment for that one-kilometre section. Based on the average
lumpsum payment for routine maintenance of $61/km/month, this works out to a payment deduction of only
$3 for that month. The total deduction for non-compliance with all performance standards for the entire length
of the road would only amount to 110% of the total lumpsum payment – it must be noted that this non-
compliance with all performance standards is almost impossible to achieve. For recently rehabilitated roads as
included in the second pilot, this means that a contractor could do absolutely nothing and likely still be entitled
to a significant portion of the monthly lumpsum payment. Although continued non-compliance can lead to
termination of the contract, such low levels of deduction percentages are undesirable. The deduction
percentages should be significantly higher, especially for important defects such as potholes (in the order of
50% rather than 5%), especially since they are applied to the payment for the one-kilometre section that is
non-compliant. The deduction percentages for the different performance standards should be reassessed for
future PBM contracts.

114. As a result of the current PBM approach, with response times to correct defects, generous performance
standards, and low deduction percentages, the penalties applied to the PBM contractors have been very low.
Where normally one would expect a low compliance and high penalties in initial months as contractors gain
experience with the approach, this has not been the case in Tajikistan. For the first pilot, the mid-term review
found that compliance with the performance standards after two years was in the order of 95.0%, while the
lumpsum payments made to the contractors over the two-year period were in the order of 99.7% (in one
contract the payment was in fact more than 100%, which should not be possible). This indicates a lower than
expected level of non-compliance with the performance standards, but especially a much lower than expected
level of applied penalties. This means that the PBM contracts are not resulting in the desired target condition
for the road, and that the payments are higher than they need be. Further improvement of the PBM approach
is required to increase its efficiency and effectiveness.

13 If the defect is not corrected in the defined time period, the deduction becomes a penalty that cannot be recovered.

27
115. The PBM contracts have been limited to routine and winter maintenance, with a provisional sum for
emergency maintenance. As mentioned earlier, the RAMS analysis is expected to result in a shift of focus
towards periodic maintenance of important roads. It is therefore recommended to expand the scope of the
PBM contracts to include periodic maintenance. This may be in the form of an output- and performance-based
road contract (OPRC) with periodic maintenance at the start of the contract paid on a volume basis, followed
by routine and winter maintenance on a performance basis. It is also possible to include periodic maintenance
at some stage during the contract period, with clearly defined triggers defining when the periodic maintenance
is warranted (based on roughness or deflection for instance). Given the expanded scope of works, it is also
recommended to expand the duration of the contracts to at least 5 years. It is strongly recommended to
develop such expanded and improved PBM contracts and to pilot these in the coming years.

8. ROAD MAP
116. This section describes the road map to be followed in improving road asset management in Tajikistan,
with a particular focus on introducing a road asset management system (RAMS) and improving the financing
of road maintenance. An overview of the different steps in the Road Map is provided in Table 23.

ROAD ASSET MANAGEMENT SYSTEM (RAMS)

117. The development of a RAMS will initially be supported under the ongoing ADB funded CAREC Corridors 2,
5, and 6 (Dushanbe–Kurgonteppa) Road Project. At the request of the government, part of the available
funding has been redirected towards the development of a RAMS. This project is expected to continue for at
least two more years, giving ample time for a consultant to be hired to support the development of the RAMS.

118. As part of the project support, survey equipment should be procured as soon as possible, allowing this to
be used by the RAMS consultant for the collection of data in at least 1,000 km of international roads in 2021
(preferably this should be expanded to include the entire international road network). The project will also
procure the necessary equipment and software for the RAMS unit. The development and initial testing of the
RAMS database and GIS mapping should be completed in 2021, and an initial HDM-4 analysis should be carried
out in the same year as the basis for preparing a 5-year rolling maintenance investment plan .

119. In 2022 the support from the RAMS consultant will be continued, to a large extent through a local partner
(preferably including a full-time presence with MOT). MOT will be responsible for collecting data on the rest
of the international road network with support from the consultant as needed. This data will be used to further
test the RAMS database and GIS mapping. An HDM-4 analysis will be repeated based on the new data and an
updated 5-year rolling maintenance investment plan will be prepared.

120. Continued support to RAMS development will be required in 2023 and 2024, and it is recommended to
include funding for this under the new ADB funded CAREC Corridors 2, 3, and 5 (Obigarm–Nurobod) Road
Project. Data collection and database management will be largely carried out by MOT with limited consultant
support. Continued consultant support will be required for the HDM-4 analysis and the preparation of the
rolling 5-year maintenance investment plans based on data for the entire international and republican road
network.

INSTITUTIONAL FRAMEWORK

121. Although considerable institutional reform has been proposed by development partners and is being
considered by MOT, this road map focuses on the institutional changes required to ensure the proper working
of the RAMS.

122. It is important that the RAMS consultants have counterpart staff to assist them and who they can involve
in the development of the RAMS, providing formal and on-the-job training throughout the process. Initially
this may involve a task force of at least 2-3 staff members from each of the three different units involved in
road maintenance planning. These staff members should be formally identified by MOT in the next couple of

28
months. Throughout 2020 the task force members will be involved in the RAMS development and receive
training from the RAMS consultant.

123. In the course of 2021, MOT will put in motion the steps required to create a formal RAMS unit with at
least 6 staff positions. MOT will also ensure that an annual budget is made available to the RAMS unit to cover
the expenses of collecting data and operating the RAMS. The budget allocation will be continued in subsequent
years.

PLANNING

124. The RAMS will need to be integrated into the existing planning and budgeting system. A detailed review
of the current planning system will be carried out by the RAMS consultant. This will be used to define how the
RAMS may be integrated into the planning procedures and budget allocation criteria, and an agreement will
be reached with MOT in this regard. The steps and responsibilities for the preparation of the maintenance
plans and related budget requests, and the role of the RAMS in this, will be described in a Road Maintenance
Planning & Budgeting Manual to be prepared by the RAMS consultant.

125. Based on the HDM-4 analysis, the RAMS consultant will also be responsible for preparing 5-year rolling
maintenance investment plans for 2022 and 2023, which will form the basis for the MOT road maintenance
budget requests for those years. This will be continued by the RAMS unit in subsequent years with limited
consultant support under the new project. Upon approval of the road maintenance budget by MOF, the RAMS
consultant will support MOT in determining the optimal allocation of the available funding to the different
roads and regions/GUSADs.

126. Apart from the HDM-4 analysis, the RAMS consultant will support MOT in identifying additional criteria
to be applied to the division of available maintenance funding between different parts of the country and the
allocation to individual roads. This may include criteria defining the division of available funding to the different
regions, but can also include criteria prioritizing roads connecting to regional or district centres.

FINANCING

127. Based on the HDM-4 analysis and the 5-year rolling maintenance investment plans, the RAMS consultant
will carry out a detailed assessment of the road maintenance needs. The RAMS consultant will also make a
more detailed assessment of the potential for financing the maintenance costs through road user charges.
Under the new project, the consultant will develop a tolling system to be piloted in at least two road sections.

128. By the end of 2021, an agreement will be reached with MOT and MOF regarding the introduction of road
user charges for financing road maintenance. This will include an agreement on the possible introduction of a
second-generation Road Fund through which to channel the revenue from these road user charges. Any
necessary legislation and related documentation for the tolling pilots will also be processed in this year.

129. The agreed road user charges will be introduced in 2022. This will involve a gradual introduction, with the
level or scope of the road user charges being gradually increased over a number of years. At least two tolling
pilots will be initiated this year.

130. By 2024, road maintenance financing will be increased to at least 0.25% of GDP (compared to 0.1%
currently). This will be the result of a combination of the introduction of road user charges and their allocation
to road maintenance, but also the increased allocation from the Republican Budget.

IMPLEMENTATION

131. The implementation will aim to increase the involvement of the private sector and the execution of
periodic maintenance by means of competitive bidding. It will build on the experiences with the performance-
based maintenance pilots, expanding the approach to include periodic maintenance and extending the
contract duration to 5 years.

29
132. Under the new project a detailed review will be carried out of the PBM experiences in 2020 while the
current PBM contracts are still ongoing. This will aim to improve the performance-based components of the
contracts, avoiding the use of response times, improving the performance standards and applying appropriate
deduction rates. The review will also look how the contracts may be expanded to include periodic maintenance
at some point in time during contract implementation. Suitable road sections for piloting the new PBM
contracts will also be identified.

133. A detailed analysis of the selected road sections will be carried out in the spring of 2021, allowing the
bidding documents to be finalized. Procurement will be done in the summer with the contracts to start
before the winter. A mid-term review will be carried out towards the end of 2024.

30
 Table 23 Road Map
2020 2021 2022 2023 2024
• Road survey equipment • Data collection carried out for • Data for remaining • Data for republican roads • Condition data for at least 80%
procured at least 1,000 km of international roads collected collected of international roads repeated
international roads
Road Asset
Management
• RAMS equipment and software • RAMS database and GIS • RAMS database and GIS
System
procured mapping developed mapping reviewed

• HDM4 analysis carried out • HDM4 analysis carried out • HDM4 analysis carried out • HDM4 analysis carried out
• RAMS task force identified • RAMS task force members • RAMS unit created and staffed
trained in data collection,
management and analysis
Institutional • Budget for data collection • Budget for data collection • Budget for data collection
Framework allocated allocated allocated

• Detailed analysis of current • Procedures for RAMS • RAMS results used as basis for • RAMS results used as basis for • RAMS results used as basis for
planning procedures carried integration in planning and budget request and allocation maintenance plan maintenance plan
out budget allocation agreed

Planning • 5-year rolling maintenance • 5-year rolling maintenance • 5-year rolling maintenance • 5-year rolling maintenance
investment plan prepared investment plan prepared investment plan prepared investment plan prepared

• Road network statistics • Road network statistics

published annually published annually
• Detailed analysis of • Introduction of road user • Agreed road user charges • Road maintenance funding
maintenance funding needs charges agreed with MOF introduced increased to at least 0.25% of
and road user charges carried GDP
out
Financing
• Tolling system developed and • Required legislation for tolling • Tolling pilots initiated in at
potential road sections for pilots in place least two locations
tolling identified
• Detailed review of PBM pilots • Identification of suitable • Incorporation of best practices • Mid-term review of new PBM
locations for expanded PBM and periodic maintenance into contracts
contracts including periodic PBM bidding documents
maintenance
Implementation
• Procurement of new PBM
contracts

31