CHAPTER III

Vendor Managed Inventory

– An Overview
3.1 Introduction

The term “Inventory” originates from the French word, Inventaire and the

Latin word Inventariom indicating a list of things found. The more popular meaning

of Inventory includes materials – raw, in process, finished packaging, spares and

others stocked in order to meet an unexpected demand or distribution in future. It

includes:

i. Raw Materials, Components and Fuel are inputs that are stored in advance for

industrial & production processes so that they function smoothly.

ii. Work-in-progress are goods that cannot be sold in the market in their present

form but awaits processing, packaging or further operations in the production

process.
iii. Finished goods are stocks that exist at the end of the production and the value-

addition process and are ready for final use by the market.

iv. Consumables are also called Maintenance, Repair & Operating (MRO)

inventories and supplies consumed during the production process but which

do not become part of the final product. These include lubricating oil, machine

spares, soap etc.

Merchandise meant for resale is not usually included under the definition of

Inventory.

The increasing challenges encountered by all organizations across sectors in

the management and control of inventory necessitated an efficient inventory

management innovations. It was in the 20th century that analytical techniques were

developed to improve inventory management practices.

Each of the different components of inventories has its own relevance in the

supply chain. If the production system is functioning in batches or just-in-time,

investment in raw materials is not desirable. For organizations engaged in mass

production, it is not economically feasible to order raw material inventories every

time production is scheduled. Sometimes the nature of the product prohibits

overinvestment in finished goods inventory, as in the case of expensive goods and

perishable products.
3.2 Classification of inventory based on the functionality in the production

process

Movement Inventories or Transit Inventories or Pipeline Inventories

These are materials like coal and fuel, which do not serve any purpose while

they are being transported to plants or factories and hence end up as inventories.

Buffer Inventories

Buffer inventories are excess stocks that are maintained to meet the

unexpected surges in demand that may arise due to unforeseen situations in the

market. These inventories are also maintained by manufacturers to tide over scarcity

in the supply of raw materials.

Anticipation Inventories

These are held if a future demand of a product is anticipated. For example,

crackers before Diwali, umbrellas before the rain, inventories in a grocery store

before a strike.

Decoupling Inventories

When different machines and people work at different speeds, stoppages in at

some sections on the shop floor are likely, which interrupts the production

temporarily. Companies keep decoupling inventories at each machine so that

production is not hindered. This ensures a smooth production process in the event of

a machine breakdown during a job.

Cycle Inventories

Usually purchases are made in larger quantities than the required level to

leverage on ordering cost.

3.3 Components of Inventory Cost

Inventory in an organisation involves different components of costs that

constitute its entire inventory cost. Inventory costs can build up whether inventories

are held or not. The different components of inventory costs are explained in detail

below.

Purchase Cost or Production Cost

Manufacturing organizations incur production costs whereas purchase cost is

incurred when the items that are bought from outside sources. Therefore, it is referred

to as Nominal cost of inventory.

Ordering Cost or Set-up Cost or Procurement Cost

Ordering cost is the cost incurred when inventory is replenished. It includes

the costs associated with the processing and chasing of the purchase order,

transportation costs, inspection costs and costs of expediting overdue orders. It is also
called Procurement Costs. If the item is produced within the organisation, there are

costs incurred in developing the production schedules and in getting the system ready

for production. The unit ordering cost or set-up cost declines as the purchase

order/production run increases in size. The total ordering cost or set-up cost is

independent of the order size and is quoted “per order”.

Carrying Costs or Storage Costs or Holding Costs

This component represents the cost that is associated with storing of an item in

the inventory and is proportional to the quantity and the time over which the

inventory is held. It includes the opportunity cost of the capital invested in the stock,

the costs directly included in storing the goods (like salary of storekeeper, rates of

storage, heating, ventilation, safety costs and racking), the obsolescence costs of

scrapping and reworking, deterioration costs and costs in preventing deterioration

and fire & general insurance costs.

Stock-out Costs

This is the cost of not serving the customer due to shortage in supply. Internal

stock-outs implies loss in production resulting in idle time for men and machines. It

can also lead to imposition of penalties on delayed deliveries. External stock-out

poses a threat of loss in potential sales and customer goodwill due to frequent back-

orders.
3.4 Reasons for underinvestment in inventories

Gaither & Frazier (2002) are of the opinion that even though inventories are

indispensable for efficient and effective operations of a system, there are

justifications for not having inventories. Some of the justifications are mentioned

below

i. A huge burden of carrying costs has to be incurred if inventories are carried by

an organisation. This will include all the costs incurred to insure, finance,

store, handle and manage large inventories.

ii. The time required to produce and deliver customer orders is increased if large

in-process inventories clog the production system. This diminishes our ability

to respond to changes in customer orders.

iii. As large inventories clog the process, more resources are needed to clear the

congestion and coordinate the schedules.

iv. Inventories are assets and l44arge inventories reduce the return on investment

thereby adding to the costs of firm.

v. Inventory also represents a form of dead investment. Materials that are

ordered, held or produced before they are actually needed result in wasted

production capacity.

Total inventory costs are crucial in taking inventory investment decisions. The

relationship between the different cost-components involved in Inventory

Management is shown in Figure 3.1.

 Fig 3.1 – Relationship between Order Quantity and Inventory Costs

It may be noted that the total cost decreases with the increase in the order

quantity, initially then starts increasing. The order quantity (Q) for which the total

cost is minimum is known as the Economic Order Quantity (EOQ) and is expressed

in terms of units or rupee value of units ordered. In case of production, EOQ is

referred as Economic Batch Size.

3.5 Mathematical models of inventory management

As organizations strive to control their inventory costs, over the years different

mathematical models have been formulated to facilitate informed decision making in

handling inventories under varying production and purchase situations. A few of the

popular models are mentioned below.

a) Classical EOQ Model or Wilson-Harris Model

This is the most elementary model of all the inventory models and is an

analytical one. This model holds true under the following assumptions

i. Demand for the item is certain, continuous and constant over time.

ii. Lead time is the time between placing an order and its delivery, is known and

fixed. When the lead time = 0, the delivery of the item is instantaneous.

iii. Within the range of quantities to be ordered, the “per unit holding cost” and

“ordering cost per order” are constant and thus independent of the quantity

ordered.

iv. The purchase price of the item is constant. No discounts are available on

purchases of large lots.

v. Inventory is replenished immediately as soon as the stock level reaches zero.

No stock shortages or overages are there.

As shown in Figure 3.2, suppose, we begin with a stock of Q on time 0. This

will be consumed at the rate of d units per day. If the stock can be replenished

immediately as soon as the stock level reaches zero, then the lead time = 0. A fresh

order is made and inventory obtained at T 1. We place an order at point A when we

have a stock equal to the demand required during the lead time LT. This is then

consumed by the time the fresh delivery is due to arrive. It arrives at T 1. The order is
placed when the inventory is at R = Reorder level. The interval between two

successive points when orders are placed, (i.e.) A and B is called the Inventory

Cycle. It is the time elapsed in consuming Q and is denoted by T.

Fig 3.2 – Inventory Profile of EOQ Model or Wilson – Harris Model

Maximum Inventory held = Q

Minimum Inventory held = 0 then,

Avg Inventory Level = (Q+0)/2 = Q/2.

 As per the assumptions, the purchase price of the units is uniform, no need for

safety stocks, decisions as to when to order and quantity to order are known. So, we

consider only the ordering cost and the holding cost.

Let Q = Ordering Quantity

O (Q) = Total annual ordering cost

H (Q) = Total annual holding cost

T (Q) = Total (Variable) Annual Inventory Cost.

Therefore, T (Q) = O (Q) + H (Q)

If N = No of times an order is placed in a year

C0 = Ordering cost per order

Then O(Q) = NC0

If annual demand = D and order quantity = Q,

then N = Orders in a year =D/Q

and O(Q) = (DC0)/Q

If Ch = Unit Holding cost (in Rupees per unit per year)

and Q/2 = Average Inventory Held, then H(Q) = (QCh)/2

Therefore, T(Q) = {(DC0)/Q} + {(QCh)/2}

To find the value of Q corresponding to the lowest value of T (Q) we find the

derivative and equate it to Zero.

d /d (Q) of {T (Q)} = d /d(Q) of {(DC0)/Q} + {(QCh)/2} = 0

Solving we get - (DC0)/Q2 = Ch/2

=> Q = [(2DC0)/Ch]1/2 = Q* = Economic Order Quantity or EOQ

To verify that this point is the minimum point, the second derivative > 0.

Here d2/d (Q) of {T (Q)} = [(2DC0)/Q3] which is a value > 0.

If T (Q*) = Minimum Annual Inventory Cost

Then T (Q*) = {(DC0)/Q*} + {(Q*Ch)/2}

Substituting Q* = [(2DC0)/Ch]1/2 in the above equation, we get

T (Q*) = [2DC0Ch]1/2

From the above explanations,

T* = Q*/D and

N* = 1/ T* = D/ Q*

Where T*= Optimal Interval between successive orders = Inventory Cycle Time

N*= Optimal number of orders placed in a year.

b) EOQ Model with Price Breaks

In real life, the assumption that unit costs of the item under consideration is

uniform may not be practical. Price discounts are based on the quantity for which

orders are placed. Lower rates for larger orders are quoted and one or more price

breaks may be offered by the seller. If unit cost price was uniform, then the

purchasing cost of the item was irrelevant as far as the cost model to determine EOQ
was concerned. Under conditions of price breaks, the item cost becomes part of the T

(Q) function as follows:

T (Q) = O (Q) + H (Q) + Purchase Cost

T (Q) = [{(DC0)/q} + {(QCh)/2} + CiD]

Where Ci = C0 if Q < q1

= C1 if q1≤ Q ≤ q2

= C2 if q2≤ Q ≤ q3

and so on till

Ci = Cn-1 if Q ≥ qn

The cost curve for the model also gets modified with a break down in price and is as

shown in Figure 3.3

Fig 3.3 – Cost Curve of EOQ Model with Price Breaks

c) EOQ Model for Production Runs or The Build-Up Model

Here, the goods are received for inventory at a constant rate over time and

they are also being consumed at a constant rate. This is relevant for situations when

items are produced internally in the factory and not bought from suppliers. Inventory

builds up when the rate at which the items are produced is higher than the rate of

their usage or depletion and is shown in Figure 3.4.

Fig 3.4 – Inventory Profile of Build Up Model or EOQ Model for Production Run

Here, the total order quantity Q is produced over a period t p and is defined by

the production rate, p. The average inventory level is determined not only by the lot

size Q, but also by the production rate, p and depletion rate, d.

Average Inventory tp = Q/p

Where Q = Order quantity

p = rate of production

tp = duration of the production run.

Inventory builds up at a rate of (p – d) per unit of time.

Maximum Inventory Level = tp (p – d)

or tp = (Max Inventory level) /(p – d)

Average Inventory Level = {tp (p – d)}/2

Since tp = Q/p, then average inventory level = {Q (p –d)}/2p

Average Inventory Level = (Q/2) [1 – (d/p)]

Ordering Cost = (DC0)/Q

and Holding Cost = Ch x Average Inventory Level

Holding Cost = {(QCh)/2} {1 – (d/p)}

T(Q) = {(DC0)/2} + {(QCh)/2} {1 – (d/p)}

To find the Q that gives minimum total cost,

d/dQ of [T (Q)] = 0
[d/dQ of {(DC0)/2}] + [d/dQ of {(QCh)/2} {1 – (d/p)}] = 0

Solving we get

Q = Q* = EOQ = [{2DC0/Ch}]1/2 x [{p/(p –d)}]1/2

T (Q*) = [{2DC0Ch}]1/2 x [{(p –d)/p}]1/2

d) Inventory Model with Planned Shortages

Shortages or stock outs are generally undesirable as it means a loss of

customer goodwill, reduction of future orders and changes in market share, to name a

few consequences. In some case, customers may not withdraw the orders but wait

until the next shipment arrives. This is called Backordering. Shortages may even be

deliberately planned to occur especially when the value of the item in question is

very high and having high holding cost. So the decision will be to set off the cost of

shortages against the savings in holding costs. ‘S’ below the zero level indicates

negative inventory, (i.e.), number of units that are backordered. As soon as the lot of

Q items is received, the customers whose orders are pending are supplied their

requirements immediately and as such,

Fig 3.5 – Inventory Profile of Model IV – Planned Shortages Model

Maximum Inventory Level = (Q – S)

Inventory Cycle T is divided into two phases

t1 = Time when inventory is on hand and orders are filled as and when

they occur.

t2 = Time when it is a stock out situation and orders are placed

as Back orders.

T (Q) = O (Q) + H (Q) + S (Q) where S (Q) = Annual Shortage cost.

From the figure,

O (Q) = (DC0)/Q =>Annual Ordering Cost

H (Q) = {(Q – S)/2}Cht1 during a cycle

(Q – S) is sufficient to last a period t1

(Q – S) = t1d => From the Figure

where d = Usage rate

Q is sufficient to last a Full cycle where T = t1 + t2

Q = Td

Combining the two equations, we get

{(Q – S)/Q} = t1/T

So, t1 = T {(Q – S)/Q}

Holding Cost during a given cycle T = {(Q – S)/2}Cht1

Substituting the value of t1 in the equation for T, we get

T = [{(Q – S)2ChT}/(2Q)]

If there are N orders in a year, there are N cycles a year,

Annual Holding Cost = N x [{(Q – S)2ChT}/(2Q)]

But NT = 1 year.

Annual Holding Cost = [{(Q – S)2Ch}/(2Q)]

Shortage cost of a given cycle

= Average Shortage x Cost of Backorders per unit x Cycle Time

= (S/2) x (Cs) x (t2)

From the figure,

Q = Td So, d = Q/T

S = t 2d So, d = S/t2

So, (Q/T) = (S/t2) Therefore, t2 = TS/Q

Shortage cost of a given cycle = (S/2) x (Cs) x {(TS)/Q}

 = {(CsTS2)/(2Q)}

There are N cycles a year and NT = 1year

Annual Shortage Cost = S (Q) = [(CsS2)/(2Q)]

Therefore the annual inventory cost

T (Q) = {(DC0)/Q} + [{(Q – S)2Ch}/(2Q)] + {(CsS2)/(2Q)}

To determine Q and S values that would minimise T (Q) in the above equations, we
obtain the two partial derivatives and set them both to zero.

[δ/ δ (S)] of T (Q) = 0

[δ/ δ (Q)] of T (Q) = 0

T (Q) = {(DC0)/Q} + [{(Q2 – 2QS + S2) Ch}/(2Q)] + {(CsS2)/(2Q)}

Expanding this equation we get,

T (Q) ={(DC0)/Q}+{(QCh)/2}–{(SCh)+{(S2Ch)/(2Q)} + {(CsS2)/(2Q)}

[δ/ δ (S)] of T (Q) = [( - Ch) + S{(Ch + Cs)/Q}] = 0

Ch = S {(Ch + Cs)/Q}

So, S = (QCh)/(Ch + Cs)

[δ/ δ (Q)] of T (Q) = {( - DC0)/Q2}+ (Ch/2) – [{S2 (Ch+ Cs)}/(2Q2)] = 0

Substituting the value of S that we got from the previous equation, we get

Q* = EOQ = [(2DC0)/Ch]1/2 x [(Cs + Ch)/Cs]1/2

Substituting this value of Q* in the equation S = (QCh)/(Ch + Cs), we get

S* = [{(2DC0Ch}1/2]/[Cs2 + ChCs]1/2
Substituting these two values in the original T(Q) equation we get

T (Q) = [{(2DC0Ch}1/2] x [(Cs)/(Cs + Ch)]1/2

Besides these deterministic models, there are also a number of probabilistic

models available with parameters like Safety Stock and Service Level included in the

analysis. It is on the basis of these models that the current literature on inventory

management and control has been formalised.

Even though the theoretical models of inventory are all linear in nature, in

reality, the inventory models do not always follow linearity. An example of a realistic

inventory profile is shown in Figure 3.6.

In a realistic inventory model, the demand is not continuous and uniform. It is

discrete and irregular. In the above figure, four cycles are displayed. In the first cycle,

the demand during lead-time (DDLT) is more than the expected demand, but there is

sufficient safety stock (SS) to meet the demand. In the second cycle, the demand is so

great that the amount of SS cannot meet it and this results in a stock-out situation.

The third cycle is a condition where the DDLT is less than the expected demand. The

fourth cycle illustrates a situation where the DDLT just matches the expected

demand. Even when the demand is irregular and unequal, the amount of average

stock held would be nearly (Q/2 + SS). Notice that the DDLT would be sometimes

more and sometimes less that the expected demand. Therefore, an amount equal to

the SS would nearly always be carried. The idea of keeping safety stock is clearly to
prevent stock-outs and it is the amount of stock that the organisation would always

preserve to meet extraordinary situations.

Fig 3.6 – A Realistic Inventory Profile

Higher safety stocks are kept in situations where

i. Costs of stock-outs are large

ii. Higher level of service is sought (Meeting greater proportion of demand)

iii. Significant variations are observed in the lead-time or in lead-time demand

iv. Holding costs are smaller.

Larger safety stocks also imply larger holding costs and therefore the

management has to strike a balance between the two. Inventory Management

ensures that total inventory costs are kept to the minimum, responsiveness towards

the customer becomes maximum and deals with the following questions:

i. How much is to be ordered?

ii. How often should orders be placed?

iii. When should the order be placed?

iv. How much stock is to be kept as Safety Stock?

The accuracy of the decisions on Inventory Management is directly

proportional to the accuracy of the forecasting methods used and the relevance of the

data that is received about the consumer demand. Hence, information plays a big role

in proper management of inventories. The company’s performance is judged on the

basis of its efficiency of operations and its responsiveness to the consumer.

To ensure that the best returns are achieved on capital and other factors of

production invested in the stock, a balance should be struck between balancing the

service levels with the cost of providing a particular level of service through proper

inventory control. The obvious solution to problems concerning level of service and

responsiveness is to increase the stock levels. This will reduce back-orders, out-of-

stock situations, lead-times and an increase in customer service. Maintaining higher

levels of stock increases the related costs of carrying inventory, insurance, taxes,

obsolescence, warehousing and materials handling. Even though lower stocks reduce

the costs of inventory, lack of readily available products on demand can mar the

goodwill of the organisation.

Kapoor and Kansal (2003) define Inventory control as a set of policies and

procedures by which an organisation determines which materials it will hold in the

stock and the quantity each one of them it will carry. Efforts are made to keep the

safety stock at the lowest level possible. It is also kept proportional to the uncertainty
or risk level. Selective Inventory Control is a methodology based on the concept of

80-20 used in Welfare Economics by a German Economist, Pareto.

Using the same principle, one can find that 20% of total stock contributes to

80% of the value. This 20% of total stock is crucial to the companies as far as

production is concerned. In selective control, inventory of high value items is

controlled because they contribute to give highest returns. Lesser care is assigned for

low value items, as returns are low.

ABC Control is the basic method of Inventory Control in which all the stored

items are analysed in terms of their annual consumptions in rupees. The inventory is

then divided into three categories:

i. Nearly 10% of items account for about 80% of the total annual consumption –

Category A

ii. About 20% of items account for 15% of the total annual consumption –

Category B.

iii. 70% of the items will cover only 5% of annual consumption –Category C.

Therefore, the time, effort and expense involved in controlling A items will

have to be much higher than those for B and C category items. Based on this

principle, the pattern of control in ABC Analysis is shown in Table 3.1.

Table 3.1 – Pattern of Control in ABC Analysis

Category A Category B Category C

 Tight Control on Stocks Moderate Control Loose Control
Centralised Procurement of exact Procurement of fairly Procurement based on
quantities from multiple source. exact quantities estimated usage
Continuous check on schedule and Broad check on Hardly any check
revision wherever called for schedules required
Very low or if possible, no buffer Exact requirement No restrictions
stock needed. should be worked out
Regular follow up Some follow up No follow up
Very strict consumption Control on Not required
consumption is highly
desirable
Materials planning forecasts Past consumption may Rough estimates are
should be very accurate be taken as the basis satisfactory.
Source: Kapoor & Kansal (2003)

Other selective control methods are also used and their selection is based on

the criteria followed by different companies vis-a-vis inventory control. Eight of the

most commonly used selective inventory control methods are summarised in Table

3.2 given below. The decision on “how much” to stock is based on the data regarding

demand rates and lead-time. The accuracy of the decisions on Inventory Management

is directly proportional to the accuracy of the forecasting methods that are used and

also the relevance of the data that is received with respect to the consumer.

Table 3.2: Various Selective Inventory Control Systems

Methods Stands for Criteria of Usage

A-B-C Always-Better-Control Annual value of consumption of
 the items concerned
H-M-L High-Medium-Low Unit price of the materials
V-E-D Vital-Essential-Desirable Critical nature of the component
in production
S-D-E Scarce-Difficult to Obtain-Easy to Obtain Purchasing problems with
respect to availability
G-O-L-F Government-Ordinary-Local-Foreign Source of the material
F-S-N Fast Moving-Slow Moving-Non Moving Frequency of Issues from the
stores
S-OS Seasonal-Off Seasonal Seasonal factor in Usage
Source: Kapoor & Kansal (2003)

Hence, information plays a vital role in proper management of inventories.

Chopra & Meindl say that the company’s performance is judged on the basis of its

efficiency of operations and its responsiveness to the consumer. Proper utilisation of

ready and relevant information makes a company more responsive and efficient

resulting in better forecasts, reduced inventories and more customer-centric

decisions. Proper use of information regarding the demand is pivotal to this study as

real-time information exchange across the supply chain of a company is the basis of

Vendor Managed Inventories (VMI). VMI systems have been developed after a

series of successful innovations over the years.

Kulkarni & Sharma (2005) describe the beginnings of series of innovations

just after the World War II. The Executive Vice President at Toyota, a Japanese

citizen Taiichi Ohno who travelled in 1956 to the United States to visit a few

automobile plants, provided the much needed boost to Inventory Management during
the middle of the 20th century. The ‘Supermarkets’ in America caught his eye. Ohno

was amazed by the way in which shoppers selected ‘what’ and ‘how much’ they

wanted all by themselves. Later this materialised into a ‘Pull System’ at Toyota, in

which each production line became a ‘supermarket’ for the next in line. Each line

would make available only the item that the next line wanted and nothing more. This

was possible by the kanban (sign board) system which indicates to the preceding

stage what the next stage wanted as replenishment of components and subassemblies.

This was improved upon by the kaizen system (continuous improvement),

made famous by Masaaki Imai, which stressed on making the right item the first

time. The Japanese even invented machines that shut down as soon as an error in

production is detected so that defective items do not pile up. All these were done with

the intention of reduction of costs due to bad production practices.

Defects were detected using a Cause-Effect Diagram called Ishikawa diagram

(after its creator, Kaoru Ishikawa) or the fishbone diagram, which was a pictorial

display of the possible causes of problems, their origin and the possible replacement

strategies.

During this time, the quality expert, W. Edwards Deming was in Japan

conducting Statistical Quality Control programmes at their production plants. He was

instrumental in popularising the Total Quality Management (TQM) philosophy

among the Japanese. TQM encompassed every aspect of the management of a

company and quality control was not practiced just on the shop floor, but also in the
administrative offices. In time, by the 1970s, Japanese products flooded the

American markets and a glaring difference in the quality as well as cost of goods was

noted. Americans too started replicating the Japanese success story and the Japanese

production system of kanban-kaizan-ishikawa was repackaged for the developed

world as Just-In-Time (JIT) Production in the later years.

Also known as Lean or Stockless production, JIT aims at eliminating sources

of manufacturing waste by producing the right component at the right place at the

right time. An attempt is made to work towards the goal of achieving the “ideal lot

size of one unit” and “a queue size of zero.” As a cushion against possible problems,

underutilised capacity of the plant is used whenever an excess requirement arises

instead of blocking capital in buffer inventory. This results in an improvement of

product quality and a significant reduction in delivery lead-time, breakdowns and

other costs of machine set-up.

A uniform load is given to all work centres through constant daily, repetitive

production. Single digit set-up times are aimed at through better planning, product &

process redesign. Suppliers encourage more frequent deliveries so that economic lot

sizes can be manufactured and inventory kept to a minimum. Production lead-time is

also kept to a minimum by reducing queue length, moving workstations closer,

application of group & cellular layout.

To cut down on unnecessary transportation costs, suppliers are encouraged to

relocate closer to the factory. Preventive maintenance schedules are followed to

avoid machine idle time. Job rotation and cross training are practiced by the work

force so that there is a smooth substitution in case of absenteeism. Workers are given

more responsibility for the quality of the work they do so that they can stop work

whenever a defect is observed. This was called the 'Jidoka’ programme.

Supplier quality assurance is stressed upon and a zero defect programme

implemented. Techniques like ‘JIT Lights’ were used to indicate slow work and

stoppages. The Kanban system of using cards & bins to convey parts between

workstations in small quantities (ideally, one unit) were used, though it is not

compulsory for a JIT programme. A Kanban programme, on the other hand can be

implemented only when JIT is being used.

Cudmore (2003) describes a method called Postponement as being based on

the principle of seeking to design products using common components, platforms or

modules but in which the final assembly takes place only when the final market

destination or customer requirement is known. To have full advantage of the system,

products and processes are designed and engineered in a way that semi-finished

products can be assembled and configured to provide high levels of variety.

Inventory is held only at the generic level and not at the finished goods level. Mass

customisation is possible. There is high flexibility and forecasting at the generic level

is easier.

Flexible Manufacturing Systems or FMS tried to meet the customer

requirement of additional variety of products in shorter and shorter time frames. This
brought about a fundamental shift away from economies of scale to economies of

scope, (i.e.), smaller quantities of a wider range of products. A minimum quantity of

every variant of a product is produced daily so that a minimum stock of every item is

always available. The machines are also so designed that they can be reconfigured

within no time to make the next product. This is practiced by Dell Computers who

are the world’s largest custom-built PC manufacturers. It achieves a strategic fit with

their competitive strategy of providing any configuration asked by the customer in a

few days time. This ensured that very few inventories were kept in stock.

The concept of JIT has transformed the world of Logistics. To make a

company responsive and efficient, a responsive and efficient logistical support is

imperative. It is to ensure that the demand is captured real-time and the right product

is delivered at the right place at the right time. This led to companies using new

concepts in logistics like 3PL (Third Party Logistics) and 4PL (Fourth Party

Logistics).

3PL involves a specialised logistics service provider who has entered into a

contract for a given period with the supplier or shipper. Kulkarni and Sharma (2005)

say that they provide transport, warehousing, freight payment and audit, inventory

and value added services like reengineering of the supply chain with respect to

logistics. Several multinationals like Fed Ex, UPS, and DHL Express are 3PL players .

A 4PL provider is a supply chain integrator that assembles and manages the

resources, capabilities and technology of its own organisation and that of

complementary service providers to deliver a comprehensive supply chain solution. It

is a combination of management consulting and 3PL.

JIT tried to eliminate inventory and allowed buyers and suppliers to have a

coordinated work environment. JIT II or Lance Dixon Bose Configuration was a

modification on JIT which helped to eliminate the buyer and salesman. It empowered

the supplier to post an employee inside the buyer’s production facility and write his

own purchase order, thus giving complete freedom to the supplier to make key

decisions about the supplied product. This was the pre-cursor to the now popular

Vendor Managed Inventory (VMI) system.

3.6 Vendor Managed Inventory

The American Production and Inventory Control Society defines Vendor

Managed Inventory (VMI) as a means of optimising supply chain performance in

which the supplier has access to the customer’s inventory data and is responsible for

maintaining the inventory level required by the customer. It is accomplished by a

process in which re-supply is done by the vendor through regularly scheduled

reviews of on-site inventory.

VMI is the first approach that allows information to be used more

intelligently. This concept is finding its applicability in automobile, apparel and retail

sectors mainly due to the advancements in the field of information technology with

broader bandwidth and gadgets like Bar Code Scanners, Electronic Point of Sales
(EPOS) data capture equipments, Electronic Data Interchange (EDI), Radio

Frequency Identification Devices (RFID) and mainly the Internet. Over the last two

decades, we can clearly see a slow transition in the power of managing a supply

chain shifting from the retailer to the vendor. Chopra et al (2000) calls it as a change

of convenience.

VMI is referred in many ways based on sector of application, ownership

issues and scope of implementation. Lee et al (2000) termed it Quick Response (QR).

Cachon & Fisher (1997) referred to it as Synchronized Consumer Response (SCR),

Continuous Replenishment (CR) or Efficient Consumer Response (ECR). Holmstrom

et al (2000) referred to it as Rapid Replenishment (RR) and Collaborative Planning,

Forecasting and Replenishment (CPFR).

The concept of VMI evolved mainly due to the rising costs experienced

through the supply chain. The errors inherent in the forecasting methods used in a

supply chain will manifest itself in every subsequent process along the chain. This

results in inventories that are either above or below the required levels, both of which

can prove to be dangerous to the company in the long run.

3.7 Bullwhip Effect

The term “bullwhip effect” coined by Lee et al. (1997a,b) refers to an

economic condition relating to materials and product supply and demand. The
bullwhip effect creates large swings in demand on the supply chain resulting from

relatively small but unplanned variations in consumer demand that escalates with

each link in the supply chain. It is a situation where the information about the final

customer's actual demand is distorted from one end of the supply chain to the other.

The errors in demand forecasting causes the Bullwhip Effect (Lee &

Padmanabhan, 1997) which is a phenomenon due to which the size of the inventory

‘shortages’ or ‘excesses’ rise, the further a firm is from the final consumer in a

supply chain. The time lags together with use of batch orders from the buyers

amplifies demand fluctuations as they go up the supply chain and makes the demand

variations more at the supplier’s end than at the buyer’s end.

A bullwhip has a structure where the end that hits the bull is thin (analogous to

actual demand) and the handle of the whip which is held is much thicker than the

cracking end (analogous to the wrongly forecasted demand which is much higher

than the actual demand). Events that trigger the bullwhip effect include increases or

decreases in order frequency and order quantities, batching of orders to reduce the

shipping costs, price reductions or sales quotas as part of sales promotions at the

store, return and refund policies, all of which can create differences between

forecasted figures and actual consumer demand. These triggers can happen at any

point in the supply chain – consumer, retailer, distributor, manufacturer or raw

material supplier. A slight variation at one link gets amplified as the order moves up

the supply chain.

 Each level visualizes a greater demand than it actually seeks to fulfil. Once the

amplified need reaches the top of the supply chain, an oscillation is triggered, which

swings the supply and demand variables in the opposite direction. This triggers an

oversupply that drives down demand, with the net result being excess inventory

storage costs, excess inventory procurement costs, lost profits, and substandard

service at each level.

Variations also happen due to uncontrollable reasons like weather, natural

disasters, industrial accidents and fires, strikes, changes in the tax rates and other

political reasons. A recent example can be cited where the Toyota Production System

famed for its Just In Time production ran into rough weather globally due to the

Japanese earthquake and tsunami resulting in demand-supply variations all over the

world, thus necessitating production to be run from places like India, to meet the

variation in demand.

Farther the supply chain partner from the final consumer, more distorted and

amplified is the error in the forecast. Lee et al. (1997a,b) identified that there are five

causes of bullwhip - non-zero lead-times, demand signalling processing, price

variations due to promotions, rationing & gaming and order batching.

The traditional supply chain is a system where each supply chain partner bases

the production or delivery orders solely on the basis of sales to the customer,

inventory levels and at times work-in progress (WIP) targets. Each stage in the

supply chain has information only on what the customers prefer and not on which

products the end customer will actually buy on that day. This lack of visibility of real
demand leads to a lot of guessing which in turn causes problems in the supply chain

if it is not properly designed. This situation is called the Forrester effect.

Houlihan effect is also called rationing effect. As shortages or missed

deliveries occur in the traditional supply chain, customers overload their orders,

leading to bullwhip effect. Some organizations resort to batch ordering to achieve

economies of scale but this can lead to bullwhip effect as actual requirements are not

reconciled. The error in forecasts caused through batch ordering is called Burbridge

effect.

Industrial engineers and economists have over the years tried to reduce

inventories at each stage along the supply chain by capturing the demand position at

each stage more accurately than before. This results in better forecasting and

operational efficiencies and this is achieved through proper management of the four

supply chain drivers – facilities, transport, inventory and information.

To save the costs of financing inventory, companies at times enter into VMI

arrangements just to force their suppliers to manage inventory for the entire supply

chain and shift this cost to them. It was observed in many cases that suppliers simply

took ownership of the inventory without major changes in the management styles.

Holding this inventory increased supplier’s costs and they had to raise their prices

which resulted in no real reduction in costs to the subsequent supply chain members,

as they merely shifted costs back and forth the supply chain.
 As Chopra et al (2000) mentioned, attempts to optimise the performance by

the individual links in the supply chain results in the sub-optimisation of the overall

supply chain. Cost reduction can be achieved only if the inventory carried is

positioned in the supply chain at a point where the supply chain effectiveness is

maximised.

Most literature like Lee & Padmanabhan (1997a & 1997b) show that the

unwanted Bull Whip effect can be reduced through advanced information sharing

through the supply chain, which in turn leads to lower inventory levels and better

supply chain performances. Lin and Shao (2000) have found that the sharing of

information in a retailer-supplier partnership is critical to both the parties. Cooray

and Ranatunga (2001) studied the power relationships between the retailer and

supplier and found out a cultural angle to it.

Over the decades, the companies that have implemented VMI have seen more

accurate transfers of information across the supply chain due to the advancements in

Information Technology (IT), Radio Frequency Identification Devices (RFID),

Barcode readers & scanners and Electronic Data Interchange (EDI). This has led to

the capture of demand information as close to the consumer as possible and in very

accurate forecasts. This in turn has reduced the inventory levels throughout the

supply chain.

A typical VMI programme involves a supplier who monitors inventory levels

at the customer’s warehouses using information sharing technology and assumes the
responsibilities of tracking and replenishing that inventory. Aparajit (2005) says that

the basic objective of VMI is to give the manufacturer access to Point of Sales (POS)

information to set the replenishment activity closer to actual sale and also to give the

manufacturer a demand-centric view of replenishment and production planning.

VMI is a backward replenishment programme that uses the exchange of

information between the retailer and the supplier to allow the supplier to manage and

replenish the merchandise at the store or warehouse level. The buyer is charged for

the material only when it is used, so the vendor incurs the carrying cost of stocking

the material in the customer’s facility. The purchase orders are created on the basis

of the demand information exchanged by the retailer/customer. This makes the

manufacturer more customer-centric as the demand data is captured closer to real

time. This also provides improved visibility of information across the supply chain

pipeline and results in fewer stock-outs and greater cost reductions.

In a VMI system, the onus of forecasting and inventory management shifts to

the vendor/supplier/manufacturer (V/S/M) and is not with the

retailer/distributor/customer (R/D/C). EDI is to be established and EPOS systems

have to be set up so that real time sales and inventory data is transferred to the

V/S/M. The supplier creates purchase orders based on the inventory data and fill

rates. Fill rates are the proportion of customer demand that can be satisfied from the

available inventory. The vendor creates and maintains the inventory stock plan for

the retailer. The vendor sends the Advance Shipping Notices before the product is
sent to the retailer’s warehouse. Then, the vendor follows this up with an Invoice to

the retailer. On receipt of the product or consignment, the retailer does invoice

matching and handles the payment. Kumar & Kumar (2001) has illustrated this as

shown in Figure 3.7.

 Figure 3.7 – INFORMATION
RETAIL AND PRODUCT FLOW DIAGRAM
IN A SIMPLE VMI MODEL
STORE
Product
RETAILER’S
INVENTORY
Product Activity Product Excess Products WAREHOUSE

& SALES VENDOR

HISTORY
• Updates Inventory Plans
Data Sent • Forecasts retailer’s demand
• Reviews Suggestions from the
retailers
• Issues Advance Shipping
Notices and Invoices.
• Ships the products
Invoice Matched, Payment Sent • Collects excess products sent Product
back from retailer warehouse
RETAILER Invoice notices sent

OFFICE

Source: Modified from Kumar & Kumar (2001)

Since VMI uses automatic electronic messages to keep track of the current stock

situation and planned sales forecasts, it also leads to improved inventory turnovers

and lower safety stock. It has been found that ever since Wal-Mart took up the VMI

initiative, the textile industry also has adapted VMI into their operations. Ironically,

the industries that face complex situations are among the last to adopt this.

Supermarkets and malls haves also started adopting this though a little late than

expected. Auto component manufacturers in India have taken this up in earnest too.

A very high level of trust and confidence is necessary between the parties.

Long-term contracts are typically entered into with VMI systems. Yao et al (2005)

mentions that VMI is a collaborative venture that authorises the suppliers to manage

the buyer’s inventory of stock-keeping units (SKU). Customers or retailers no longer

place orders, but instead only share information with the vendor, who takes

responsibility for inventory management. No explicit orders are received by the

vendor, instead an indication of the upper and lower stock limits that are expected to

be kept are intimated. Within these stock-bands, it is the supplier’s responsibility to

forecast and replenish inventory.

VMI is a ‘pull’ replenishment practice which represents the highest level of

partnership where the vendor is the primary decision maker in the order placement

and inventory control processes. Simchi-Livi et al (2000) in their study mentions that

the supplier decides an appropriate inventory level for each product within previously

agreed boundaries and appropriate policies to maintain these levels.

 VMI is a reality in the 21st century due to the advancements in the information

technology and telecommunications sector. High speed, high quantity data transfer is

today possible due to leased line and broadband technologies. Immediate and

elaborate data capture is also possible due to barcode scanners, EPOS (Electronic

Point of Sales) Retrieval systems and EDI (Electronic Data Interchange) devices.

Accurate tracking of any tagged device on this planet is now possible with RFID

(Radio Frequency Identification Devices) technology. Internet helps buyers share

sales and inventory information on a real time basis with suppliers, who use several

analysis software to plan production runs, scheduling deliveries, deciding

transportation routes, balancing the load of vehicles and production facilities and also

manage the order volumes and inventory at the buyer’s SKU facilities.

VMI offers the authority to that supply chain partner who has the best

visibility of the inventory position. It uses information more intelligently than in the

traditional inventory systems. VMI uses effective real time sharing of POS data

through EDI systems and analytical software tools to calculate the appropriate order

quantities, product mixes, travel routes and safety stock.

Thus VMI pushes the supply chain decision-making responsibility up the

chain. The v/s/m is in a better position to support the objectives of the entire

integrated supply chain. A wider range of customers and their demand patterns can

be observed resulting in improved forecasting. The effects of promotional

programmes and seasonal changes are automatically appended to the POS data thus

giving a better forecast.

3.8 Models of VMI

i) Authority Transfer Model

Here the inventory management cost and other activities regarding the

inventory are transferred to the vendors by the buying organisation. In its simplest

form, the supplier's representative visits the customer’s premises, to count the

inventory so as to get a picture of the replenishment requirements. Here, though the

customer gets benefits by shifting the responsibilities of inventory management to the

supplier, the whole supply chain costs remains the same. In this model, either the

costs are transferred back to the buying organisation or it eats into the operating

profits of the supplier.

ii) Joint Planning Model

This collaborative planning model consists of two stages – first being the data

sharing stage and the second takes care of forecasts and production schedules that get

jointly developed by the supply chain partners. The ‘buyer’ collaborates with the

supplier on demand and consumption details so that an agreed upon consensus

forecast of future demand is reached which favours the profits of both the companies.
iii) Fully Automated Replenishment Model

This model incorporates the positives of both the above models by aiming to

reduce total supply chain cost by focussing on costs at both buyers’ and supplier’s

premises. At the macro level, it defines the objectives and limitations of the

relationships and the power equations. At micro level, it develops the replenishment

strategy for each SKU. Once the replenishment strategy is developed, the daily

demand and the inventory level at the replenished site are closely monitored via EDI

and all strategies are based on that data.

It is this fully automated replenishment model that is currently the blueprint

for the existing VMI systems. The other two models were modified forms of JIT and

JIT II.

3.9 Global scenario of VMI

V.K. Magupu, whole-time director and Sr. Vice President, IT & Technology

Services, L & T and Chief Executive, L & T Infotech in his keynote address as Chief

Guest at the VMI Conference jointly organised by L&T Infotech and the Indian

Institute of Materials Management in Mumbai in Nov 2004 had said,

“VMI is the next big thing after Just In Time and it has tremendous

potential to create real win-win situations for both the buyer and

vendor. VMI enforces true partnerships between the buyer and vendor;
 cleans up an organisation’s processes and establishes an ecosystem,

with VMI being the base motivator for improving efficiency”.

The tie-up between North America’s largest retailer, Wal-Mart and their

suppliers, Proctor & Gamble (P & G) is widely regarded as the first VMI venture. P

& G received sales data directly from the checkout counters which was used for

planning and scheduling of production and delivery to the Wal-Mart stores on a

replenishment basis.

VMI has existed much prior to EDI and other technologies. Much before the

Wal-Mart- P&G arrangement, Frito-Lays delivery trucks replenished retailer’s

shelves without going through the store owners as the involvement from the

management of the stores was found to be non-value adding for them. Even the

practice of JIT II where the supplier posts their representative at the retailer's store is

also “Vendor Managed.”

Lee & Padmanabhan (1997a, 1997b) in their path breaking research paper on

VMI have cited a number of examples worldwide to highlight the usage and results

of VMI implementation, some of which are cited below.

Wal-Mart and P&G had a VMI program together for more than ten years to

manage the inventory and production of disposable diapers, which was later extended

to other suppliers and products. Major retailers like Wal-Mart, K-Mart, Dillard

Department Stores and JC Penney, all have VMI agreements with their vendors.
Their experiences are that high volume, high turnover items are the best candidates

for VMI though it is arguable.

Chopra et al (2000) also cites a few examples in their research works. At K-

Mart, inventory turnover on seasonal items have dramatically increased to more than

10 times and for non-seasonal items to as high as 17-20 times. The customer service

levels have also risen higher than usual after VMI implementation. K-Mart has

anyway scaled back their VMI program to just 50 suppliers from an initial high of

300 confirming Dr. Chopra’s claim that VMI is best with few big vendors.

Grand Union, a New Jersey based grocery retailer with more than 100 stores

and three Distribution Centres improved inventory turnover by close to 80% and

achieved 99% service levels. Warehousing costs and out-of-stock conditions had also

decreased. Grand Union entered into VMI only with vendors who were proven

market leaders in their category and who were comfortable with technology and

supply chain innovations.

ACE Hardware, the largest hardware cooperative has also seen fill rates rise

from 4% to 96% in the past few years after VMI was introduced. Oshawa Foods, a $6

billion Canadian food distributor and retailer had VMI arrangements with Pillsbury,

Quaker and H.J.Heinz which ran into a lot of trouble initially due to hasty

implementation. Once the system was fine-tuned, customer service levels of 99%

were achieved and inventory turns multiplied 3 to 9 times.

 Fred Meyer, the 131-unit chain of super centres in the Pacific Northwest

region after its VMI with two main food vendors improved service levels to 98% and

reduced inventories by 30% to 40%.

Hughes (1996) mentions that Panduit, which is one of the world’s largest

electrical component manufacturers (with about 60,000 SKUs in stock), did not see

cost savings in the VMI system in the initial stages. The company markets

exclusively through 1800 distributors in the US, who in turn sell to maintenance &

repair shops, construction firms. Profits were very less due to the heavy competition

in this field. EDI was initially used only for entering orders and not for generating

purchase order acknowledgements, advance shipping notices and invoicing all of

which were done by fax. Since all these were not integrated into a single system, the

full benefits of VMI could not materialise. Later, it developed a turnkey VMI/EDI

system called Qualified Supplier Program (QSP) created by an external vendor,

Advantis for a tailored approach specific to its industry. Panduit persuaded 34 major

suppliers to adopt QSP. Now, Panduit knows exactly what every distributor carries in

their warehouses at any point of time and automatically ships the needed items along

with generating an electronic invoice and advance shipping notice. Whenever the

customers are seen to be maintaining too much stock, Panduit generates a return

order with no restocking charges. So, they now claim to have started providing better

service and out of stock conditions are almost non-existent. This highlights that
industry specific tailoring is needed for certain systems to achieve the actual potential

of VMI.

Cataldo (1996) mentions that Motorola Inc’s semiconductor sector has a VMI

system which uses EDI to provide online visibility of a customer’s inventory and to

forecast data that provides inputs to the planning systems. Six of Motorola’s OEM

(Original Equipment Manufacturers) customers are now using this system which

allows Motorola to determine the quantity of product that a customer needs, time of

shipping and the requisite quantity. Motorola uses an auto scheduling tool known as

The Order Promising System (TOPS) and a shipment tie-up with logistics giant, UPS

called Rapid-NET in congruence with the VMI and has reduced procurement time

from an average of 23 days to 2 days for every execution cycle. Shipments are

handled exclusively by UPS and customers are given a choice to select delivery time

ranging from 1-4 days. Same rate is charged regardless of place of product origin.

Motorola prepays the freight, bills the customer as a separate invoice charge thus

reducing administrative delays.

Carbone (2005) mentions that the OEM provider, Celestica Inc., partners with

key suppliers to provide lowest cost solutions to its customers using VMI and by

stocking at locations that are close to manufacturing facilities, most of which are

outsourced to Electronics Manufacturing Service providers in China, Taiwan etc.

Michaeu (2005) mentions a case about Boeing’s Skin and Spar facility that

has a VMI relationship with Alcoa, its raw material supplier for the plant that makes
large wing products. VMI arrangement has gone a long way in solving the

forecasting and delivery problems. As a result, Boeing started sending weekly

electronic forecasts, weekly electronic inventory counts and purchase orders to

Alcoa. Alcoa in turn changed their order entry process to include the VMI agreement

and using this data for internal production. This has led to a strong and flexible

supply chain, more accurate forecasts and better production decisions.

Lenius (2005) mentions about Elkay which implemented a VMI for water

cooler and fountain products with Indianapolis based Central Supply. VMI is said to

have reduced overstock and labour related to tracking inventories there.

Sucher & McManus (2002) suggest that to streamline its supply chain

operations, Herman Miller, a leader in the industrial furniture industry, has

significantly benefited from employing a supply centre for components, managed by

the component suppliers.

All over the world, VMI is popular among raw material and component

suppliers to electronics industry, electrical sector, construction & aviation industry

and automobile giants. Retailing, grocery and apparel sectors are the other areas

where VMI is slowly becoming the industry norm. In the global scenario, the visible

benefits are reduced stock-outs, overstocking, forecasting errors, better production

scheduling and inventory turns.

 Ismim Scouras, Executive Editor of EBN, an online business magazine is of

the opinion that VMI is suitable during times of oversupply as it is easier for

suppliers to control inventory when in plenty. Jennifer Baljko, a renowned columnist

on supply chain trends has maintained throughout that there is still a disparity

between the benefits that OEMs and their suppliers receive out of VMI. In a survey

by Electronic Supply Chain Association and Chainlink Research Inc customers have

reported costs dropping after VMI implementation, while only 11% suppliers said

that their costs have reduced. They say that the initial costs involved in the setting up

of infrastructure has increased their cost of doing business and that they are forced to

enter into VMI agreements only because their customers are forcing them into VMI

to continue in business.

Vendor Managed Inventory implementation has been facilitated by the

improvements in EDI and data capture technology. RFID and its increasing use have

gone a long way in making VMI a feasible option.

3.10 Radio Frequency Identification Devices (RFID)

Mathur (2006) defined RFID as an automatic identification device technology

that is used to remotely store and retrieve data without actual scanning of the data

source. The predecessor to this technology was the bar code scanner used at retail
cash counters which needs actual line of sight scanning to read the data and bill the

product.

In 1946, Léon Theremin invented an espionage device for the erstwhile Soviet

Union, which used radio waves and had applications as a secret listening device. It

has been recognized to be the first device similar to RFID technology. This

technology was introduced in a paper by Harry Stockman (1948).

The first true ancestor of modern RFID was Mario Cardullo’s U.S. Patent

3713148 in 1973 which was demonstrated in 1971 to the New York Port Authority

with applications as a toll and traffic detection device at the ports. The first patent to

be associated with the abbreviation RFID was granted to Charles Walton in 1983 as

U.S. Patent 4384288.

RFID was introduced to initially improve the Supply chain but has found

applications in manufacturing, retailing, warehouse traffic management, military,

medical & healthcare, education sector and e-governance. The companies in India

taking to this technology are steadily on the rise and it has the capability to transform

the business equations.

An RFID system comprises of an RFID Tag or transponder, RFID

transceivers, high capacity servers and related application software. An RFID chip

consists of a tiny computer chip, which is approximately the size of a small dot, on

which are implanted, the code of the product and a small antenna.
 RFID can incorporate a variety of electronic architecture and code formats. To

bring in standardization, especially in the retail sector, a code format called EPC

(Electronic Product Code) has been proposed by EPCglobal (which was earlier called

as Auto-ID Centre).

Arora (2006) studied that the generally accepted EPC based RFID format is a

result of a collaborative research work done by Auto-ID Centre, MIT and over 100

huge corporations that included Wal-Mart, US Department of Defense, US Food and

Drug Administration, US Postal Service, Pfizer, Coca Cola, Philips, Microsoft,

Infosys Technologies and IBM Consulting.

The image of an RFID Tag and the Electronic Product Code Structure is shown

below

Fig 3.8 – RFID Tag with the EPC Code and Structure

Source: Auto ID Centre

RFID Tags can be active or passive. Active RFID tags have a limited operating life

and are powered by an internal battery that power the chips to generate signals giving

longer reading range. Active RFIDs are of comparatively larger size costing more

than the passive RFIDs. Passive RFID tags operate without an external power source

by using just the power generated from the reader and the incoming radio signal.

These are lighter than active tags, less expensive, more widely used, have shorter

reading range and have almost unlimited operating life. These have been mentioned

in the paper by Saxena & Doctor (2006).

Low frequency RFID Systems are used in the 30KHz to 500KHz range and

high frequency systems are used in 850MHz to 950MHz and 2.4 GHz to 2.5 GHz

range. The data transmitted can provide identification information, location

information, the product details like batch number, colour, date of purchase, shelf

life, time on shelf till now, price, date of manufacture, time spent in transit, location

of distribution centre, name of last person to hold the item along the supply chain

among other details depending on the level of information required on the tag for

different product categories.

3.11 Key application features of RFID

Raza, Bradshaw & Hague (1999) studied the several application features of

RFID devices when compared to the earlier data storage, retrieval and transaction
processing devices like Bar Code scanners. RFID tags need not be visible to be

read/scanned. The tags can be read quickly from significant distances. A number of

tagged devices can be simultaneously read at a time. As most of the tags come

enclosed in a protective covering, it is difficult to tamper them in normal situations.

Since it can be encased in protective covering, they can be protected from harsh

environments, fluid & chemicals and rough handling. Many tags now come with both

read and write capabilities, rather than just read-only facilitating addition of

information after some significant event in the movement of the tagged item takes

place.

In 1998, a group of researchers at Massachusetts Institute of Technology

(MIT) decided to find a solution for interoperability issues and standardization of the

use of RFID so that related hardware and software costs could be brought down. The

MIT Auto-ID centre which did pioneering research in this field found out that the key

to reducing the cost of RFID Technology was to focus on reducing the functionality

on the tag. A unique identifier called Electronic Product Code (EPC) was devised

which acts as an ID that point to more detailed information of the tagged item. Global

standards also allow interoperability and increase the adoption of the technology

thereby reducing the costs.

In 2003, the intellectual property of MIT Auto ID Centre was transferred to a

joint venture of European Article Numbering International (EANI) and Uniform

Code Council (UCC), which was called as GS1 (Global Standards 1). EPCGlobal
which is a subsidiary of GS1 develops and oversees the implementation of these

standards for the EPC Network. EPCGlobal also maintains the EPC number registry.

The EPC is embedded in an RFID tag and is remotely transferred to the system, thus

reducing the cumbersome work of opening boxes and scanning using barcode

scanners.

3.12 RFID and its role in reduction of Bullwhip Effect

Bullwhip effect causes a lack of visibility into the supply chain by distorting

that replenishment stock data thereby resulting in a wrongly placed order. The

solution for this is a visibility into the supply chain, which modern technology like

RFID enables. Each link in the supply chain has its own ordering routines, reorder

points, and quantities. Ordering to forecasts, rather than to actual need, distorts

inventory.

A company's estimation of its inventory may deviate from the actual

requirement in the market thus creating a shortage or a surplus. A retailer may also

inflate its orders during a shortage, only to find that a shortage was short-lived.

The strongest contributing cause to the bullwhip effect according to Simchi-

Levi & Kaminsky (2000) is a lack of centralized demand information. Each link in

the supply chain uses its own method of estimating inventory instead of using actual
customer demand as a measure. Centralized demand information thus becomes the

solution. But seamless connectivity along a supply chain is rare.

Both Wal-Mart and the Department of Defense (DoD), USA have attempted

to reduce the bullwhip effect by mandating the use of RFID by their suppliers.

Members include Wal-Mart, the DoD, and Procter & Gamble, and hundreds of others

suppliers. Wal-Mart uses RFID to take precise inventories; it then holds its suppliers

accountable to deliver the exact replenishment. The suppliers in turn are required to

tag cases and pallets of the inventory they supply, so that Wal-Mart can conduct a

precise incoming inventory audit. Boeing and the DoD are beginning to mandate

RFID visibility into the shop floors of some key suppliers. RFID follows the work

order, enabling Boeing to see the exact status of a given order.

RFID, along with global positioning systems (GPS) and similar technologies,

are contributing to "The Internet of Things", a term coined by Kevin Ashton, who

was once a brand manager at Procter & Gamble and later co-founded the RFID

company, Thing Magic. The Internet of Things or RFID gives items a unique identity

(much like a URL) and uses the Internet to transmit information about it, such as its

whereabouts and condition. With a well-run Internet of Things, using technologies

like RFID to make items searchable, stock-outs, overstocks, and the bullwhip effect

can (in theory) be vastly reduced.

3.13 Global applications of RFID in retail & Supply Chain Management

Wal-Mart has introduced RFID attached to each pallet and storage box that

comes into/goes out of their stores and distribution centres and almost completely

replaced bar codes. In June 2003, Wal-Mart had communicated to its major suppliers

that in two years, all pallets and boxes should come tagged with RFID. Information

about the contents loaded onto a roller or box can be entered onto the tag and easily

checked. This helps to check if some materials are missing during transport. Also, at

warehouses, a common mistake committed is when a box is loaded into the wrong

loading bay and eventually into the wrong vehicle. By the time, this error is detected,

rectification becomes late, especially in cases of perishable goods. RFID alerts the

warehouse officials by being connected to an alarm system when wrong items are

loaded to the wrong loading area. It can also be incorporated to a receiving station

where a wrongly dispatched item can be identified if delivered wrongly.

Friedman (2005) mentions that the use of RFID at Wal-Mart store has reached

a stage where Wal-Mart can identify those products that move faster on each

weekdays. The purchase patterns of different demographic consumer groups can be

separately analysed. The system can alert the store manager when the temperature at

which the perishable goods are stored in the refrigerator varies from the set limit.

Nowadays, retailers all over the world are tagging their products and the level

of pilferage has come down. Earlier, shoplifting was rampant in busy supermarkets

but now, any unbilled item automatically sets off an alarm at all the exit points.
Retailers are tagging child trolleys of shoppers so that they do not encounter “missing

children” situations and can ensure child security.

The uniqueness of the RFID tags mean that the product can be individually

tracked as it moves from location to location, finally ending up in the customer’s

hands. This can help combat the problems of theft and product loss as mentioned

before but also have advantages in recall campaigns for products with quality

deficiencies. This can help in post-sale tracking and profiling of customers for future

campaigns too.

Throughout the European Union, RFID passes are used for the public

transport systems. This system has now been copied by Canada, Mexico, Israel,

Dubai and Columbia also. All the transport payments and toll charges are monitored

and done through RFID Compliant systems. This reduces a lot of time spent by

logistic companies along the motorway and can speed up the checking & inspection

stages in the logistics. This automatically brings down the cost of transportation.

Another major application is in animal tracking when meat and livestock are

transported throughout a country before it reaches supermarkets. The RFID

implementation helps identify the farm from which the animal has been loaded, its

date of birth, age and nutritional value along with history of any contaminations if

any. The Canadian Cattle Identification Agency began using RFID tags for its

purposes.
 RFID finds applications in bookstores and libraries for tracking its inventory.

Other applications are in airline baggage tracking, pharmaceutical items tracking,

building access control, shipping container tracking, truck and trailer tracking. The

pharmaceutical industry is highly vulnerable to counterfeiting with figures suggesting

that 7-8% of world market is counterfeit. RFID technology can help protect against

fraudulent introduction of drugs into the drug supply chain. Pfizer has already

incorporated this system to their drug supply chain.

The automotive sector introduced use of RFID by tagging the car keys. This

application ensures that the car does not start without the actual key. Toyota Avalon

2005, Lexus GS 2006, Toyota Camry 2007, Toyota Prius and companies like Ford

and Honda are introducing car models with this feature being optional. The driver

can even open the doors and start the car with just the presence of the key within

three feet without even taking it out of the pocket.

Tyre manufacturer, Michelin tested RFID embedded tyres in 2003 to offer

tyres in compliance with the United States Transportation, Recall, Enhancement,

Accountability and Documentation (TREAD) Act which aims at safer road transport

for trucks that are involved in logistics operations in the supply chain. Tyres nearing

the expiration date can be recalled ensuring safer vehicles on the highways. The

Malaysian government has introduced RFID passports for proper tracking of travel

history of its citizens.

3.14 Applications of RFID in Indian Retail & Supply Chain Management

Retailers, textiles, aviation, energy and auto sectors in India are switching to

this new concept over the last five years due to the successful implementations

elsewhere. This is also necessitated by pressures on them by suppliers from abroad to

comply with global business practices, failing which they run the risk of losing global

clients.

Pai (2006) mentions that Infosys Technologies is a founding member of EPC

and Wipro Technologies have been associated with Auto-ID Lab at MIT for some

years now. Both these companies play a big role in the EPC which provides

standards for implementation of the technology. Similarly, Gemini Traze RFID Pvt

Limited is India’s first RFID tag manufacturing unit at Sriperumbudur Electronic

Park near Chennai. It rolls out 45 million units per year which would be increased to

100 million units per year later on.

Kishore Biyani’s Future group company, Pantaloon has piloted an RFID

project at one of its warehouses in Tarapur using more than thousand RFID tags as

mentioned in Biyani & Baishya (2007). It selected a few lines of apparel for the

RFID pilot project. The application was developed by Wipro Infotech and integrated

Oracle database. Nowadays, we witness major retailers using flying saucer shaped

plastic knob like structure on dresses that are on display at apparel stores which are

removed when they are billed. This helps in tracking of goods and security from

pilferage as it lets out an alarm at the exit door if not billed properly.
 Mathur (2006) mentions about Madura Garments who also experimented with

RFID and has incorporated them in their Planet Fashion stores as well as factories

and warehouses. The national carrier Air India is planning to use RFID for tracking

capital assets. Leading oil companies have begun pilot tests to use RFID for LPG

cylinder tracking, The Indian railways is also comprehending on these lines for

tracking wagons and containers. Maruti Udyog Limited has already started using

RFIDs for component and spare parts tracking at their Gurgaon plant. Ashok Leyland

is also using this for the same purpose. It has tremendous advantages as there are

more than 20,000 parts in most vehicles and tracking the movement of each one of

them through the supply chain is a mind-blowing task. Mahindra & Mahindra are

also using RFID in some of the manufacturing processes like Pre-treatment of Body

Shell and Electro-deposition that are done in harsh conditions.

In the pharma sector, Ranbaxy Labs and Pfizer use it for counterfeit

protection. Airport Authority of India is considering RFID for the cargo and

passenger goods management. Saxena & Doctor (2006) explains about applications

outside of retail and auto sectors, citing the libraries like Jayakar Library of Pune

University and Dhanvantri Library of Jammu University who have adopted RFID.

Hyderabad Central University has introduced RFID embedded degree certificates.

The municipal corporation at Hyderabad in 2006 had introduced RFID for keeping

track of garbage collection & disposal trucks and their drivers to monitor them due to

instances of reported malpractice in collection & disposal. Applications are there in

healthcare where new born babies can be RFID tagged so that they are properly

monitored in hospitals. It can be used to improve security and in military uses for

proper tracking of supplies to the armed forces. There are applications possible in

Electricity/Water meters which can help make the manual recording and reading

automatic, remote and fault free. The applications are enormous and far-reaching,

and India has just begun. RFID tagged employees can be monitored better by

knowing their location in different areas of the retail facility. A person spending too

much time in the restroom can be alerted to go back to the store and an overworked

person can be asked to take rest.

Therefore, the possibilities of RFID implementations once the technology

stabilizes are exciting, wide ranging and multifaceted. The next chapter looks at the

review of literature and the findings of other studies conducted all over the world on

the subject of VMI Implementations.