CONSTRUCTION

INDUSTRY
INSTITUTE

PROJECT CONTROL FOR ENGINEERING

Bureau of Engineering Research Publication 6-1

The University of Texas at Austin July 1986
 The purpose of this publication is to make available to industry the results of re-
search conducted by the Construction Industry Institute (CII). The publication does not
necessarily represent the views of all CII member companies, but is offered as a contri-
bution to the industry.

CII was founded in 1983 to improve the cost effectiveness of the nation’s largest
industry. The members, who represent a broad cross-section of owners and contrac-
tors, believe that many of the problems that limit cost effectiveness are common ones,
and that real improvements can be best accomplished in a cooperative environment
with the benefits being shared by the construction industry at large.

The activities of CII include identifying research needs, conducting research, and
aiding implementation of research results. The task forces that provide guidance for
specific CII activities are listed below.

83-1 CICE Impact Evaluation

83-2 Productivity Measurements
83-2-3 Model Plant
83-3 Constructability
83-4 Industry Data and Statistics
83-5 Contracts
83-6 Cost/Schedule Controls
83-7 Materials Management
84-8 Design
84-9 Technology
84-10 Quality Management
85-11 Employee Effectiveness
85-12 Project Organization
85-13 Safety
85-14 Education and Training

The Construction Industry Institute

The University of Texas at Austin
3208 Red River
Suite 300
Austin, Texas 78705-2650
(512) 471-4319
Project Control for Engineering

Prepared by
The Construction Industry Institute
Cost/Schedule Controls Task Force

Publication 6-1
July 1986
 First Printing: July 1986
Second Printing: March 1987
 TABLE OF CONTENTS
Chapter Page

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

2. Organizations for Engineering . . . . . . . . . . . . . . . . . . . . . . . 2

General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Functional Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Task Force Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Matrix Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
The Status of Project Controls in the Organization . . . . . . . . . . . . 3

3. Project Control Basics for Engineering . . . . . . . . . . . . . . . 4

Types of Engineering Contracts . . . . . . . . . . . . . . . . . . . 4
The Nature of Engineering Work . . . . . . . . . . . . . . . . . . 4
Planning and Budgeting . . . . . . . . . . . . . . . . . . . . . . . . . 5
Terminology.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Breakdown Structures and Work Packages . . . . . . . . . 7
Schedules.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Codes of Accounts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Progress Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Procurement Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Controlling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Historical Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
System Design and Computer Selection . . . . . . . . . . . . 14
DOE/DOD Cost & Schedule Control System Criteria . . . . . . . . 15

4. Other Observations and Findings . . . . . . . . . . . . . . . . . . . . . 16

Obtaining Project Participant Support . . . . . . . . . . . . . . . 16
The Key Role of the Project Manager . . . . . . . . . . . . . . . . 16
How Much Should Be Spent on Project Controls? . . . . . . . . . . . 16
Changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
What Makes Projects A Success? . . . . . . . . . . . . . . . . . . 17
The Future . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 7

5. Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19

Appendix A: Progress Measurement Systems . . . . . . . . . . . . . . . .20

Appendix B: The DOE/DOD Cost & Schedule Control

Systems Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
 1

INTRODUCTION

One of the most important challenges facing engineering and construc-

tion management today is the prevention of cost and schedule overruns of
the type that have plagued major projects so often in the past. The multiple
parties involved in a project (owner, engineer, construction manager, con-
tractors, and subcontractors), the complexities inherent in a project, the
pressure of time, increasing regulatory requirements, inflating costs of ma-
terial and labor, and the uncertainties of nature combine to make major
engineered construction a management nightmare. Obviously a project
must have a formal project control system if there is to be success, and that
system must encompass the engineering effort.

Project control is concerned with resource quantities, time, and costs.

Planning and controlling of resources is generally accomplished within the
cost control system. All resources, whether they be personnel, material,
equipment, supplies, or services, are quantifiable and can be converted to
a dollar amount, During the planning of a project, quantity and dollar
budgets for these are established. During the controlling operation, actual
quantity and dollar performance and rates of utilization are compared with
planned performance.

Time and work progress are managed through the schedule. During the
planning operation, engineering work tasks are scheduled to fit project re-
quirements and a system for measuring progress of each task is established.
During the controlling operation, status is reported against that system.

The benefits of an effective project control system are many. It docu-

ments the project plan and actual performance. It identifies problem areas
and unfavorable trends. It is a communication tool. It allows project man-
agers and other project participants to “keep a handle” on the work. It
feeds into the historical database so that future planning of comparable
work can be more accurate. Overall, the project control system is a major
factor in the success of the project at hand and the planning of future
projects.

Control of engineering activity is generally more difficult than control of

construction activity because engineering tasks are more difficult to quan-
tify and track between start and completion, tasks are more parallel and
overlapping, and the responsibility for a given design often is shared
among various disciplines. The challenge is particularly great on those
projects in which engineering overlaps procurement and construction. En-
gineering work, however, can be controlled. Succeeding chapters of this
publication provide details of a system for this control.

1
 2

ORGANIZATIONS FOR ENGINEERING

General

A properly structured organization is key to effective management and

control of any project. The planning and controlling of the engineering for
a project, where multiple functional departments and many individuals are
involved, is particularly complex. It is essential that the work included
within the project be fully scoped and defined, and that responsibility be
assigned for its accomplishment. Three types of organizational structures
may be utilized for managing this work - a functional organization, a task
force organization, or a matrix organization.

Functional Organization

A functional organization is the traditional pyramidal organization

wherein each functional department (e.g., Engineering, Accounting, Con-
struction) is organizationally independent of the others and handles all
work within its area of expertise. Coordination between departments is
handled on an informal basis and resolution of conflicts falls to the senior
managers to whom the various departments ultimately report. This type of
organization works well when a minimal amount of contact is needed be-
tween departments. Within a functional organization, project management
becomes an additional duty of senior company managers. It is a feasible,
though not particularly efficient, organizational arrangement for manage-
ment of small projects; it is not effective for large projects.

Task Force Organization

A task force is an organization comprised of individuals assigned from

the functional departments to the task force for the purpose of managing a
project. Individuals assigned to the task force are committed full time to
the project during the period of their assignment. A project manager is
appointed and the other personnel are organized functionally under the
project manager. The organization and personnel of the task force may
change as the project moves through various phases. The major advan-
tages of a task force are that it permits concentration of resources on a
project and involves minimum management levels. This type of organiza-
tion is particularly suited for large projects.

2
Matrix Organization

The matrix organization is a combination of a task force organization

and a functional organization. In a matrix, there are two chains of com-
mand for each staff member, one along functional lines and the other
along project lines. This is achieved by having the staff report to two man-
agers simultaneously - a project manager for project direction and a func-
tional manager for technical direction. The project team within the matrix
is a group whose members are drawn from various functional departments
of the organization and who are committed to support the project or task at
hand. Project (team) managers are normally drawn from a pool of such
managers. In some situations, both project managers and team members
may simultaneously serve on more than one project team.

There are strong and weak matrix organizations. A strong matrix is one
where the project team structure is preeminent - i.e., the project manag-
er’s authority over the individuals on the project team exceeds that of the
functional managers. In a weak matrix, the reverse is true.

Matrix organizations are the most common and also the most complex
form of organization utilized for management of engineering work. When
well managed, they provide for maximum coordination, information ex-
change, and resource sharing. Matrix organizations are also the most eco-
nomical form for management of multiple projects, none of which is large
enough to warrant a full task force. The strong matrix is the preferred
form 2 .

The Status of Project Controls in the Organization

Project control, to be effective, must begin during the planning and bud-
geting of the work. It will continue during the conceptual and detailed
engineering phases. If these phases are a part of a contract where engineer-
ing, procurement, and construction overlap, engineering project control
will be integrated with procurement and construction project control.

The project control function is the “eyes and ears” of management at all
levels and the source of project status information for the client. It is also
an information center for every professional on the staff. Accordingly, it
should be organizationally placed so that it responds directly to the project
manager. It must not be treated as another accounting function, nor should
its activity be decentralized among the functional groups. And it must be
recognized as an integral part of management, not as a police force! Man-
agement must assess the information provided, then take the necessary
steps to implement the action needed.

3
 3

PROJECT CONTROL BASICS FOR ENGINEERING

Types of Engineering Contracts

In industrial construction it is common for engineering, procurement,

and construction to be handled in parallel and with some overlap - these
are called Engineering-Procurement-Construction (E-P-C) projects for pur-
poses of this publication. The engineering portion of the project may be
part of a turnkey arrangement administered within a single agency or it
may be handled as a separate contract that must tie-in with the contracts of
others.

It is possible, however, for engineering to precede completely both pro-

curement and construction. An engineering contract in this case is referred
to in this publication as Engineering-Only (E-O). The E-P-C format is the
more challenging of the two from a project controls viewpoint.

The Nature of Engineering Work

The primary products of engineering work are documents to guide both

the construction of a facility and the manufacture of engineered equipment
for installation in such a facility. Other products are studies, procedure and
operating manuals, and various consulting services.

In performing its work, the engineering group is generally organized

around engineering disciplines (civil, architectural, structural, mechanical,
electrical, etc.) or specialty services (model shop, procurement, environ-
mental studies, economic studies, etc.). These production units of the or-
ganization have products such as drawings, specifications, models,
manuals, studies, and procurements. Supporting them are such functions
as word processing, computer services, accounting, personnel, and others.

Some owners maintain internal organizations for accomplishment of

conceptual and/or detail engineering of their facilities. But much if not
most of it is handled under contract with an engineering firm. Many con-
tracts in the latter case are reimbursable, although fixed-price contracts are
becoming more common, and there are reimbursable contracts with ceil-
ings, targets, or incentives. The typical reimbursable contract is handled in
this fashion:

- Professionals within the production units are identified as reimburs-

able. The client agrees to pay for each hour of work of those reim-
bursable individuals in an amount equal to the base hourly pay of
each individual times a multiplier. This multiplier is in the range of
1.5-2.5.

4
 - Other items are identified as “reimbursable directs.” These are
items such as travel, reproduction, computer services, and commu-
nications. In most cases, their costs are passed through to the client
without markup.

- Items such as general management, accounting, personnel services,

and secretarial services are “non-reimbursable directs and
indirects,” and are not charged directly to the client. These costs
are accounted for within the multiplier noted above. That multiplier
also accounts for profit (unless the project is cost-plus) and for per-
sonnel fringe benefit costs (burden) such as vacation, medical,
FICA, and retirement.

If the contract form is fixed-price or other than the reimbursable de-

scribed above, the engineering agency still must account fully for the direct
costs, project level indirects, general overhead, and any contingency and
profit in the pricing.

Planning and Budgeting

As noted previously, there are activities of direct work in an engineering

contract and each has a cost. Other activities that are support in nature,
both direct and indirect, are only cost items. Contract progress is tracked
against the work activities; costs are tracked against all activities and other
cost items. Thus, the project control system must be designed to control
both work and cost.

When planning any project, it is desirable to divide the project into well-
defined, manageable parts for purposes of control. By controlling the indi-
vidual parts, it is possible to establish control of the total project. Figure 1
is a matrix representation of the budget that might be established for a
typical engineering project. Certain items on this budget include alloca-
tions of work-hours as well as costs. Others are budgeted only for cost.
Experienced engineering organizations are able to estimate quite accu-
rately the number of engineering documents and the work-hours required.
They are also able to estimate the quantity and work-hour requirements for
their other products and the costs of other reimbursables and non-
reimbursables and, having done this, to establish a multiplier. These esti-
mates become the basis for project control.

Control of work is best based on control of budgeted work-hours for

individual work activities (see discussion in following paragraphs for exam-
ples). As part of this control, there must be some method for evaluating
percent complete of each activity. The methods are established during the
planning operation. Methods available are summarized in Appendix A.

5
6
Terminology

Remaining discussion will describe a variety of terms used in the con-

trolling process. Figure 2 may prove helpful in explaining and interrelating
these terms. The example uses a simple process system composed of two
vessels and interconnecting piping. It traces the flow of engineering work
and relates that to the components of the completed process system.

Breakdown Structures and Work Packages

The Cost Breakdown Structure (CBS) for the project includes every ele-
ment of cost for the project. In effect, it is a summary of the budget by
category. The total of all categories equals the total project budget. Refer-
ring again to Figure 1, this matrix is a visual representation of the CBS for
an example project.

The Work Breakdown Structure (WBS) for the project relates only to the
engineering work products that the project manager chooses to status as a
basis for progress reporting. Drawings are traditionally a status item. De-
signing, specification writing, study preparation, model construction, and
equipment/material procurement involve deliverables, and thus are other
logical choices. The more of these tasks chosen for control, the more posi-
tive and informative the control system will be.

On Figure 1, the shaded portions are the budgets for those elements of
the CBS which will comprise the WBS for this example project. In this
example, only Design & Drawings, Specs, Procurement Support, and Field
Support transactions have been chosen for work control. The activities ac-
counted for within these budgets are essentially the greatest level of detail
in the WBS. Thus, the WBS is incorporated within the CBS. Actually, the
complete WBS for the project will have the total facility to be designed as
the ultimate summary level. The facility is then divided into well-defined
areas, systems, and structures which in turn can be further sub-divided into
well-defined sub-areas, sub-systems, and sub-structures to form a hierar-
chical WBS describing the total facility to be constructed. The WBS is con-
tinued beyond that point to identify those activities required to prepare
documents that are used to guide both the construction of a facility and the
manufacture of engineered equipment for installation in such a facility. A
work package is any one set of these activities which are targeted on a
given procurement or construction need date.

A contractor may include selected indirect activities such as supervision

or project controls in a progress control system, and use the Ratio or Judge-
ment method of reporting status (see Appendix A). In such a case, the
CBS/WBS relationship will be more like that shown in Figure 3. However,
indirect functions do not directly contribute to deliverables and inclusion
of them will distort progress on deliverables.

7
8
 What is the practice in industry on this subject? Reference 2 is a report
from a research project which provided much of the background material
for this publication. One area investigated was progress control; firms were
asked which work items were included in their progress control system. To
catalog responses, the breakdown of the total effort in an engineering
project was estimated to average 40 percent to drawing development, 15
percent to specifications development, 10 percent to design support activi-
ties, 25 percent to procurement activities, and 10 percent for other activi-
ties. With these assumptions it was found that four of six firms use only 55
percent of the total work for formal progress tracking, while one tracks 80
percent and another 90 percent.

Schedules

Different levels and types of schedules are appropriate for control of

engineering work. The most detailed schedule control occurs in the con-
trolling of individual engineering deliverables. At that level, barchart
schedules may be appropriate, but for effective schedule control the roll-
up of such tasks must be related to activities in a CPM-formatted engineer-
ing schedule.

Activities Indirects 3
Labor Equip.
Supervision $ WH $ $
internal Support $ WH $ $
Procurement $ WH $ $
Study $ WH $ $
Design $ WH $ $
Document Preparation $ WH $ $
Document Review $ WH $ $
External Support $ WH $ $
Other 4
$

Notes: 1. The CBS is composed of all activities in the matrix for which dollars ($) are
budgeted. The total value of these equal the project budget.
2. The WBS is composed of those direct labor activities in the matrix for which
work-hours (WH) are budgeted. They may be used for work progress measure-
ment.
3. “Indirects” includes supervision above first level, staff, facilities, supplies &
services, travel, etc.
4. “Other” includes office overhead, contingency reserve, profit, etc.

Figure 3. Example CBS/WBS Relationship 1,2

9
 The project’s control schedule may be in barchart form with milestones
and key interfaces on an E-O contract, but networking is certainly appro-
priate. On E-P-C projects, the control schedule should be in network for-
mat and interrelate with the procurement and construction schedules -
engineering work package completion usually constrains the start or con-
tinuation of a procurement or construction activity on E-P-C projects. Thus,
need dates for dependent work packages become milestone dates on the
engineering portion of the control schedule. On E-P-C projects it is abso-
lutely essential that a logic trail be established among all activities contrib-
uting to project completion. The trail is best established if network
relationships are created among all activities. This philosophy must be bal-
anced against having too much detail on the control schedule since it must
be simple enough to permit human review and analysis. The solution is to
keep the project’s control schedule at a network summary level and to treat
its engineering, procurement, and construction activities as hammocks
supported by individual detailed schedules. When this is accomplished
and the total system is integrated, it will be possible to trace the effects of
any changes and deviations throughout the project. As noted previously,
engineering work packages are the activities on the control schedule and
as such are the hammocks for more detailed subnets that trace both se-
quential and parallel steps in accomplishing the activities included within
the work package. More on this subject is contained in the CII publication,
Model Planning and Controlling System for Engineering-Procurement-Con-
struction of Industrial Projects.

Schedule control and updating will take several forms. Approved

changes will be the basis for changing the control schedule. Recognizing
that actual performance will seldom mirror the control schedule, control-
lers also will maintain a working schedule which documents performance
to date and plans work from that point. Short-term, detailed scheduling
will be used for planning work for the near term future (30-90 days).
Throughout the project, the control schedule will serve as the target, and
as such, must be the basis for all reporting of items such as schedule condi-
tions, progress, and others.

Codes of Accounts

A relatively complicated system of coding is needed for control of engi-

neering work. First, a two-part code will reflect the CBS matrix illustrated
in Figure 1, one part representing the function, the other the activity. Engi-
neering personnel will use this code when logging their time so that it is
charged to the proper budget category.

10
 Another numbering system(s) is appropriate for drawings, specifications,
and other deliverables. These numbers can be sequential in nature and
need not include any cross-reference to the WBS. Drawing, specification,
and other deliverable lists and control numbers are developed during the
planning and budgeting phase. The deliverables coding is needed for
cross-referencing to work packages. It is also useful for personnel working
on these items to use in charging time on their time cards. In this case it is
used in conjunction with the CBS identifier as a first step in work-hour
tracking. For example and referring to Figure 1, assume that the electrical
discipline function has the code “EL,” that the budget category is “design
and drawings” (code “01”), and the drawing number is 4567 - the cod-
ing for work on this would be:

EL-01-4567

For activities that contribute to more than one deliverable (e.g., various
criteria development), the deliverable portion of the code can be replaced
by an activity code. This numbering system is particularly useful in cost
control since it collects data at the budgeting (CBS) level. This format also
permits summary of data against function, activity, or specific deliverable
or activity type for historical purposes. It can also be incorporated in the
schedule activity identification numbers used at the detailed (sub-ham-
mock) level.

Still another coding system is needed for work packages. Since work
packages are really an extension of the WBS, the codes for work packages
are really a detailed extension of the WBS code. A typical engineering
work package includes an assortment of engineering documents and any
given document can be a component of several work packages, and thus
this numbering system should be separate from that used for included com-
ponents. This code structure is used on the engineering portion of the con-
trol schedule. Included documents will be cross-referenced to work
packages and work packages to the WBS within datasets of the integrated
control system.

The project’s WBS will have a hierarchical code structure that permits
summary of information to any level of that structure. It will be used in
conjunction with work package identifiers in scheduling.

The materials tracking system will require a coding system to identify

equipment items and commodities so that their tracking can be integrated
with the schedule. This coding system will also permit generation of equip-
ment, instrumentation, and other lists used within the control system.

11
Progress Measurement

Percent complete is determined by the method chosen during the plan-

ning operation for the task in question (See Appendix A). Earned value
techniques are used for summarizing overall work status. The earned value
of any one item being controlled is:

Earned Work-Hours = (Budgeted Work-Hours) x (Percent Complete)

Note: Budgeted Work-Hours equal original budget plus approved changes.

Overall percent complete of the project or of a work package is found by

this formula:

Sum of Earned Work-Hours of Tasks Included

Percent Complete =
Sum of Budgeted Work-Hours of Tasks Included

Trends can be tracked through various indices. The Productivity Index

(PI) and the Schedule Performance Index (SPI) are particularly useful for
this purpose. The PI provides a comparison of the number of work-hours
being spent on work tasks to the hours budgeted and is an indicator of
productivity. The formula is as follows:

Sum of Earned Work-Hours of Tasks Included

PI =
Sum of Actual Work-Hours of Tasks Included

Note: For the above to be a true indicator of productivity, only those tasks for
which budgets have been established should be included in summa-
tions.

The SPI relates the amount of work performed to the amount scheduled
to a point in time. The formula is:

Sum of Earned Work-Hours to Date

SPI =
Sum of Scheduled Work-Hours to Date

Note: Scheduled work-hours used in this formula are summarized from the
task schedules.

In both the PI and SPI formulas, an index of 1.0 or greater is favorable.

Trends can be noted by plotting both “this period” and “cumulative” PI
and SPI values on a graph.

While the SPI for the total project or for a work package is somewhat of
an indicator of schedule performance, it tells only part of the story. The SPI
only compares volume of work performed to volume of work scheduled.
There can be an SPI in excess of 1.0, but the project can still be in danger
of not meeting milestones and final completion dates if managers are ex-

12
pending effort on non-critical activity at the expense of critical activity. The
SPI does not show if work is being completed in the proper sequence.
Thus, as part of schedule control, controllers must regularly examine the
schedules of all included tasks in each work package so that any items
behind schedule can be identified and corrective action taken to bring
them back on schedule.

Cost performance on the project is tracked by comparing actual costs to

those in the budgets of the CBS. Unfortunately, significant lags between
incurrence of cost obligations and receipt of invoices or other cost notifica-
tion for many accounts can make it difficult to maintain a timely, respon-
sive cost control system. It is possible to use an earned value approach for
cost control if an owner or contractor so desires. Appendix B describes
such a system as utilized on some U.S. Government contracts.

Procurement Activity

Two types of procurement activity are associated with engineering. The

first type is the procurement of equipment/material to be engineered by
others. With this type procurement, the design engineers develop perfor-
mance specifications and other criteria for an item and provide them to the
selected vendor. The vendor designs the item to meet the specified criteria
and forwards the design documents to the engineer for review. Following
drawing approval, the vendor will fabricate and deliver the item. Some
engineering activity may have to be put on hold until approved vendor
documents are available (e.g., foundations and piping connections). Often
the timing for these vendor documents to support follow-on engineering is
more of a determinant for the schedule of early procurement train activities
than the item’s need date for construction.

A second type of procurement is the more general type which results in

the delivery of off-the-shelf pieces of equipment or commodities.

An engineering firm may have either or both types of the procurement

actions included in its contract on an E-O contract. On an E-P-C contract,
it would be normal for the firm to have both. Regardless of what agency
has procurement responsibilities, it is essential that procurement be in-
cluded in the controlling system for the project. It is also essential that
procurement activities constraining engineering design (and vice versa) be
related to the activities which are constrained.

A given procurement action is a train of sequential activities. Each of

these activities must be included in the detailed schedule, although the
overall procurement of a component can be a single activity on the control
schedule. For procurement tracking, each procurement train can be treated
as a unit and tracked using the Incremental Milestone method. Or, each
train activity can be tracked separately using a method most appropriate to
that activity (Appendix A).

13
Controlling

The system described is intended to provide the project manager and

other project participants with the information needed for control of the
project. Data generated from the system are continually analyzed for the
purpose of identifying trouble spots or unfavorable trends. If the system has
been designed properly, timely reports will be available so that corrective
action can be taken when and where needed.

Historical Data

Planning of future projects relies heavily on experience on past projects.

Datasets within the system should be designed to summarize and accumu-
late experience data in a format directly usable in future planning.

System Design and Computer Selection

Integrated project control is an objective in the selection or design of the

computerized project control support system. The following features are
desirable in the system.

- The computer software should have both a scheduling module and

database management capability.

- The schedule module must handle network type scheduling.

- The databases should be linkable to the schedule module.

- A graphics capability should be included which is capable of pro-

ducing network logic diagrams, barcharts, histograms, and graphs
from system databases.

- The system should be capable of generating user-designed reports

using data directly from any combination of the datasets, and also
should be capable of performing calculations or summations from
the data.

- Databases should be structured so that it is not necessary to enter a

piece of data more than once.

- The system should provide the project controls capabilities de-

scribed elsewhere in this publication.

14
 Improved capabilities now allow microcomputers to be used for stand-
alone project control on many projects. In choosing hardware and soft-
ware for a microsystem, users must insure that candidate systems can
handle both the scheduling and database management requirements. Con-
trol of engineering is a complicated procedure, and therefore a database
management capability of considerable capacity is required.

DOE/DOD Cost & Schedule Control Systems Criteria

The U.S. Departments of Energy (DOE) and Defense (DOD) have

adopted a policy concerning control systems used by contractors on cer-
tain types of projects. This policy is described under the Cost & Schedule
Control Systems Criteria (CSCSC) for Performance Measurement. A brief
description of CSCSC features as applied to engineering and as related to
previous sections of this document is in order. Appendix B is an overview
of the system and Reference 3 is a publication which describes the system
in detail.

15
 4

OTHER OBSERVATIONS AND FINDINGS

Recent research2 has provided a variety of other observations and find-

ings relating to project control of engineering. Several of significance are
summarized in the following paragraphs.

Obtaining Project Participant Support

In order for a project control system to be effective, it must be under-

stood and accepted by all who use it. Unfortunately, it is a fact that very
few engineering programs in the universities and colleges of the United
States include any required course material on planning and controlling
engineering work, nor are such subjects as procurement, construction
methods, and constructability included. Thus, in-house training of engi-
neering personnel in these subjects is a must if management is to expect
their understanding, commitment, and support. Additionally, individuals
providing data to the system should be given a voice in the design of input
format and the system should be designed to provide feedback to the
participants.

The Key Role of the Project Manager

Management commitment to a program is essential if a program is to

succeed. Thus, in the case of project controls, it is essential that project
managers understand the theory and potential of a modern project controls
system. Once a project manager is committed, steps can be taken to assure
understanding and commitment of all other project participants.

How Much Should Be Spent on Project Controls?

The portion of the budget which contractors spend on project controls

varies. Since every project is unique and client requirements differ, it is
difficult to determine the exact level of control needed in terms of project
budget. However, based on projects observed, it appears that good project
controls performance is best assured if at least 8 percent of the engineering
budget is committed to this effort.

16
Changes

Project control systems are based upon planned budgets and schedules.
When changes are introduced, the plans and schedules must be revised.
Some changes are the result of scope change; others result from value
engineering, constructability studies, or engineering error. Unfortunately,
changes seldom affect only one engineer or discipline; they can have a
ripple effect on the entire project. Thus, it is essential that an engineering
organization have formal scope control and change control programs.
More on this subject is contained in the CII publication, Scope Definition
and Control.4

What Makes Projects A Success?

Those interviewed were asked to identify the factors that distinguish suc-
cessful and unsuccessful projects, with the following results:

Successful Projects:
Well-defined scope
Early, extensive planning
Good leadership, management, and supervision
Involved, positive client relationship
Proper chemistry among project participants
Quick response to changes
Engineering managers concerned with total project, not
just engineering
Unsuccessful Projects:
Ill-defined scope
Poor management
Poor planning
Poor communication between engineering and construction
Unrealistic schedules and budgets
Poor personnel quality
Excessive changes
Poor project controls

The Future

Advancing technologies and new ideas can be expected to influence

project controls in the future. The following have been identified:

Computer-Aided Design (CAD). Computer-aided design has become

commonplace in engineering organizations, but systems which di-
rectly tie into estimating and project controls are in their early stages.
In the future there should be systems which automatically generate
equipment and material lists that feed into the estimating system and
resource-load the schedule while also prompting procurement activity.

17
Expert Systems. Expert systems would be valuable during conceptual
engineering to generate major equipment lists, bulk quantities, and
other information for various processes and to produce resulting cost
estimates. Such systems would also greatly simplify value engineering
studies during the conceptual engineering phase of a project and facili-
tate various decision processes throughout the detail engineering
phase. Other applications include forecasting, risk analysis, and per-
sonnel training.

Engineering Control Based on Quantities Designed. Traditional track-

ing of engineering work is based on engineering units of production.
Some firms are experimenting with a tracking system based on quanti-
ties designed (e.g., cubic yards of concrete, linear feet of pipe). Such a
system would parallel engineering control with construction control.
This approach could be automated within CAD: However, develop-
ment of good quantity estimates during the budgeting process is a ma-
jor hurdle for this approach.

Inclusion of Quality in Project Control. Time, cost, and quality con-

trol are the goals of every project manager, yet quality control is not
formally included in the project control system. Use of a quality per-
formance index along with other indices used in control systems
should be considered for the future.

18
 5

SUMMARY AND CONCLUSIONS

Effective project control of the engineering effort can be attained. Mod-

ern techniques with supporting computer systems are available, but the
industry has been slow in accepting them.

It takes organization and qualified people to manage and control com-

plex engineering work. Various organizational forms are possible, but a
task force organization is the most efficient for the larger projects; a strong
matrix management approach is more practical for smaller projects.

The project control system must be designed to control both work and
cost. The system must encompass planning, scheduling, monitoring,
reporting and analysis, forecasting, and historical data collection.
Subsystems within the total system must be available to track procurement
activity and to generate and maintain equipment lists, instrument lists, and
other summaries associated with design engineering work. The design of
the system should be based on the principle of integrated project control,
and be flexible enough to handle large and small projects while also re-
sponding to special client needs.

The entire professional staff of the engineering organization must be

committed to and support project control and be trained in the operation
of the company’s system. If properly designed and supported, such a sys-
tem should provide the control needs of management and essential
feedback to individual professional personnel.

Establishment of an effective project control system will require money

and the establishment of a formal project controls organization. Approxi-
mately 8 percent or more of the costs budgeted for an engineering project
should be allocated for project control to be effective. Eight percent may
seem high, but the potential cost and time savings attributable to an effec-
tive project control system far outweigh the cost of establishing and main-
taining that system.

19
 APPENDIX A

PROGRESS MEASUREMENT SYSTEMS

Basic Methods

The determination of percentage completion on a single activity can

usually be handled under one of the following four systems:

Units Completed. This method is suitable when the total scope of an

activity consists of a number of equal or nearly equal parts, and the
status is logically determined by counting parts completed and com-
paring that to the total number of parts included in the activity. Ideally,
each unit is of relatively short duration. While this method has consid-
erable application in the construction phase, it has limited use during
engineering. A possible application is in the writing of a number of
specifications of a given type where all specifications are considered to
have essentially equal weight.

Incremental Milestone. This method is appropriate for activities of sig-

nificant duration which are composed of easily recognized, sequential
sub-activities. Percentage completion values are established based on
the effort estimated to be required at each milestone point relative to
the total for the activity. This method is ideal for control of drawings
and can be used in procurement. A typical example for drawing con-
trol is:

Start drafting 0%
Drawn, not checked 20%
Complete for office check 35%
To owner for approval 70%
First issue 95%
Final issue* 100%

*Only when no additional engineering is anticipated

A typical example for procurement is:

Bidders list developed 5%

Inquiry documents complete 10%
Bids analyzed 20%
Contract awarded 25%
Vendor drawings submitted 45%
Vendor drawings approved 50%
Equipment shipped 90%
Equipment received 100%

20
 Start/Finish Percentages. This method is applicable to those activities
which lack readily definable intermediate milestones, and/or the
effort/time required is difficult to estimate. For these tasks, controllers
credit 20-50 percent when the activity is started and 100 percent when
finished. The reason that a percentage is assigned for starting is that
this compensates for the long period between start and finish when no
credit is being given. This method is appropriate for work such as
planning, designing, manual writing, model building, and studies. It
can also be used for specification writing.

Ratio or Judgement. This method is applicable to tasks such as project

management, constructability studies, project controls, and compara-
ble activity which involves a long period of time, has no particular end
product, and is estimated and budgeted on a bulk allocation basis
rather than on some measure of production. It can also be used on
those tasks for which the Start/Finish method is appropriate. Percent
complete at any point in time is found by dividing hours (or dollars)
spent to date by the current estimate of hours (or dollars) at comple-
tion.

Summary Levels

Determination of percent complete of summary level accounts is found

by use of earned value as discussed in Chapter 3.

21
 APPENDIX B

THE DOE/DOD
COST & SCHEDULE CONTROL SYSTEMS CRITERIA

The CSCSC System

Following are key features of the DOE/DOD Cost & Schedule Control
System Criteria (CSCSC) for Performance Measurement.

Breakdown Structures. A Work Breakdown Structure (WBS) for the

project is developed similar to that previously described in this docu-
ment. This breakdown is product-oriented and defines the columns of
a matrix. The rows of the matrix are defined by an Organization Break-
down Structure (OBS), which details in hierarchical fashion all organi-
zations responsible for the work. The rows and columns meet to
describe cost accounts which include all specific tasks to be performed
by an organization with respect to an item on the WBS. A typical cost
account might be the electrical design by Firm A of a specific system.
It can also be an indirect item such as quality control or supervision.
Cost accounts can be further broken down into specific tasks such as
drawings and specifications. The term work package is used to de-
scribe both tasks and combinations of tasks in the CSCSC system. The
CBS of Figure 1 is essentially a simplified version of the CSCSC’s
WBS/OBS matrix.

Terms. A number of terms are established for use in reporting the sta-
tus of a project. These are:

Budgeted Cost for Work Scheduled (BCWS). BCWS is the cost ac-
count or total contract budget less any “management reserve”
(funds held in reserve to be used in case of scope growth). It is
time-phased based on the cost-loaded schedule.

Budgeted Cost for Work Performed (BCWP). BCWP is the earned

value, expressed in dollars, for all work accomplished during a
period. It is equal to the sum of the budgeted costs of that work.

Actual Cost of Work Performed (ACWP). ACWP is the total of all

costs actually incurred in accomplishing work within a given pe-
riod and recorded at the cost account level.

Budget at Completion (BAC). At the cost account level, BAC is the

total authorized budget for that account. For the project, it is the
total including management reserve.

22
 Estimate at Completion (EAC). EAC is the latest cost estimate for a
cost account or the total project.

Cost Variance (CV). CV is the difference between BCWP and

ACWP, and shows whether work performed has been costing
more or less than budgeted.

Schedule Variance (SV). SV is the difference between BCWP and

BCWS, and shows whether more or less work was performed than
scheduled.

At Completion Variance (ACV). ACV is the difference between

EAC and BAC, and shows whether an overrun or underrun is
expected.

Cost Performance Index (CPI). CPI is an index found by dividing

BCWP by ACWP. It can be calculated on a “this period” or cumu-
lative basis, and when plotted on a graph is an effective way of
showing cost trends. A CPI of 1.0 or greater is favorable. It is an
alternate presentation of the CV.

Schedule Performance Index (SPI). SPI is an index found by divid-

ing BCWP by BCWS. It compares the volume of work being per-
formed to the volume scheduled. As with the CPI, it is suited for
trend charts. An index of 1.0 or greater is favorable. It is an alter-
nate presentation of the SV.

Cost Accounts. These take one of three forms:

Work Packages. Work involving easily identified end products.

Level of Effort (LOE). Support type activity which does not have a
final product and which is not specifically related to individual
work packages. Management, project controls, and other staff ac-
tivities are typical LOE cost accounts.

Apportioned Effort. Support activities which are directly tied to in-

dividual work packages. Quality inspection and testing is an ex-
ample of a cost account in this category.

23
Applicability of CSCSC to Civilian Design Projects

The structure of CSCSC is complicated and difficult to use. It was de-

signed for high-dollar value, cost-reimbursable, and complex defense and
energy projects, not the more typical commercial and industrial projects. It
monitors both cost and work through a cost control system. Unfortunately,
costs do not flow into a project in a timely manner, thus a work control
system based on cost is not timely. Most contractors prefer a work statusing
system based on work-hours and work quantities (normally reported daily)
for day-to-day management of work. If operating under CSCSC, contractors
will usually employ a work-hour/work quantity system for internal control
in addition to complying with CSCSC for reporting to DOE or DOD. The
CSCSC recognizes and allows such supplementary control systems by
contractors.

The full Organization Breakdown Structure (OBS) is another feature of

CSCSC that has marginal value in the management of civilian projects.
While the government finds such a structure useful in the management of
huge, multi-contractor, cost-reimbursable research and development
projects, contractors on civilian work generally will find that their CBS
incorporates whatever organizational identification is needed.

24
NOTES

25
NOTES

26
 REFERENCES
1. Modern Management Systems, Construction Industry Cost Effectiveness
(CICE) Project, Report A-6, The Business Roundtable, November 1982.

2. Diekmann, James E., and Thrush, K. Bryan, Project Control in Design

Engineering, University of Colorado, A Report to The Construction In-
dustry Institute, January 1986.

3. Cost & Schedule Control Systems Criteria for Contract Performance

Measurement. Implementation Guide, U.S. Department of Energy
(DOE/MA-0087), May 1980.

4. Scope Definition and Control, Publication 6-2, The Construction Indus-

try Institute, July 1986.

27
Cost/Schedule Controls Task Force Membership

Jack Browder, Brown & Root

Jerry Davis, Shell
Jim Diekmann, University of Colorado
Ron Duffield, Gulf Interstate
Dan Halpin, University of Maryland
Carl McKenry, Ford, Bacon & Davis
Bob McMakin, Union Carbide
Tom Mendel, RUST
*Jim Neil, Morrison Knudsen, Chairman
Stan Nethery, Dow
Dave Nichols, Department of Defense
Bill Schlicklin, Owens-Corning
Rich Venglik, BE&K
Tom Wilson, ALCOA

Board of Advisors Liaison

Keith Price, Morrison Knudsen

Past Members

Pat Dargan, ALCOA

Bruce Dinsmore, ARCO
Jerry Fischer, Mobil
Bob Gardner, Union Carbide
John Thiem, Kellogg-Rust

*Principal Author

Editor: Rusty Haggard

28
 The Construction Industry Institute
Member Companies

AT&T Resource Management Allstates Design & Development

Aluminum Company of America Company, Inc.
Amoco Corporation Guy F. Atkinson Company of
Anheuser-Busch Companies California
Atlantic Richfield Company BE&K Construction Company
Chevron Corporation Bechtel National, Inc.
Consolidated Edison Company of Belcan Corporation, Inc.
New York, Inc. Blount International, Ltd.
Dow Chemical U.S.A. Brown & Root, Inc.
E.I. duPont de Nemours & Company John Brown Engineers & Constructors
Exxon Research & Engineering Incorporated
Company CRS Sirrine, Inc.
FMC Corporation Cherne Contracting Corporation
General Electric Company Davy McKee Constructors Inc.
General Motors Corporation Dravo Constructors, Inc.
Houston Lighting & Power Company ENSERCH Corporation
ICI Americas Inc. Fidelity & Deposit Company of
International Business Machines Maryland
Corporation Fluor Daniel
Mobil Research & Development Ford, Bacon & Davis Incorporated
Corporation Foster Wheeler Corporation
Monsanto Company Fru-Con Corporation
Northern States Power Company GSGSB Architects, Engineers &
Owens-Corning Fiberglas Corporation Planners
PPG Industries, Inc. Gilbert/Commonwealth, Inc.
Potomac Electric Power Company Gulf Interstate Engineering Company
The Procter & Gamble Company Gulf States, Inc.
Shell Oil Company Jones Group, Inc.
Texaco Inc. The M.W. Kellogg Company
U.S. Department of Defense Peter Kiewit Sons’, Inc.
Union Carbide Corporation Lummus Crest Inc.
The University of Texas System Morrison Knudsen Corporation
North Brothers
RUST International Corporation
Santa Fe Braun Inc.
Stearns Catalytic Corporation
Stone & Webster
Turner Construction Company
Woodward-Clyde Consultants
H.B. Zachry Company

Participating Academic Institutions

Arizona State Mississippi State

Auburn New Mexico
California-Berkeley North Carolina State
Carnegie Mellon Oklahoma State
Cincinnati Oregon State
Clemson Penn State
Colorado Purdue
Colorado State Stanford
Georgia Tech Texas
Illinois Texas A&M
Iowa State Virginia Polytech
Kansas Washington
Maryland Worcester Polytech
Michigan
Not printed with state funds