Department of Systems
Engineering and Operations Research
OR 651 – Military Operations Research: Cost Analysis
Spring, 2006
Abstract: Cost Analysis is a sub-discipline of
Operations Research. While drawing on other disciplines (e.g., Managerial
Accounting, Econometrics, Systems Analysis, etc.), Cost Analysis uses the basic
tools of Operations Research to solve a specific class of public sector
problems. Cost Analysis is an inquiry to assist decision-makers in choosing
preferred future courses of action by evaluating selected alternatives on the
basis of their costs, benefits, and risks. Cost Analysis is different from Cost
Estimating in that projecting future courses of action usually requires
mathematical modeling, hence the link to Operations Research and Economic
Analysis. This course introduces the basic concepts of Cost Analysis.
This
class will follow a lecture format, but there will be a different lecturer each
week. An important topic in Cost
Analysis is scheduled for each lecture. We draw on the experiences of experts
from the cost analysis research and practitioner communities; hence, this
course provides a unique introduction to the topic. This course is particularly
useful for students of Operations Research, Public Policy, Information and
Software Systems Engineering, and Systems Engineering.
Course Coordinators: Stephen J. Balut, Director
Cost
Analysis and Research Division
Institute
for Defense Analyses
sbalut@ida.org
Andrew
Loerch, Associate Professor
Fairfax,
VA 22030-4444
aloerch@gmu.edu
Time: Thursday,
16:15–18:55
Location: Institute for Defense Analyses
Room
1301 (Note: Class will be held in room 1313 on
Prerequisites: STAT 344 (or equivalent) – Probability
and Statistics for Engineers
OR
541 or OR 542 (or equivalent) –
Introduction to Operations Research
Textbooks: Course Pack (available at the book store)
Optional texts: Quade,
E.S., Analysis for Public Decisions, 3rd
Edition.
Fisher, G.H., Cost Considerations In Systems Analysis.
Proposed Topical Outline and Lecturers:
January 26 Cost
Analysis Overview and Statistics Review
Andrew
Loerch,
Presents a concise definition of cost
analysis, its history and evolution to the present time. Cost analysis is differentiated from related
disciplines, including operations research, systems analysis and
cost-effectiveness analysis. Each
ensuing lecture topic will be described along with the contribution of the
topic to cost analysis. Those taking the
course for graduate credit will be advised of testing, grading and required
class projects. A short review of the
statistics needed for the course will be given.
February 2 Cost Effectiveness Analysis
D.
Levine, Institute for Defense Analyses
First postulated by RAND Corporation analysts
in the 1950s, cost effectiveness analysis has been accepted as the principle
analytical methodology for making resource allocation decisions in the
government. In the Department of Defense
(DoD), it is applied to choices among weapons systems, support concepts,
training methods, and many other military outputs. This lecture will highlight the
methodology: a clear statement of the
objective(s), a broadly based search for alternatives for meeting the
objective, and the quantitative analysis of the costs and effectiveness of each
alternative. Preference is indicated for
the alternative with minimum cost for achieving a desired level of
effectiveness (or maximizing effectiveness among equal cost alternatives).
February 9 Economics in Cost Analysis
S.
Horowitz, Institute for Defense Analyses
Presents the economic underpinnings of
cost analysis. In particular, the cost
function will be derived from the underlying production function relating input
and output quantities. In addition, the
rationale for discounting future costs and benefits will be given. Finally, various criteria will be compared
for choosing among alternative investment projects: cost-benefit ratio, internal rate of return,
and discounted present value.
February 16 Estimating Relationships - I: Factors and Simple Models
P.
Lurie, Institute for Defense Analyses
Introduces simple factor models and
linear regression analysis as tools for establishing cost estimating
relationships (CERs). Topics to be
covered include measures of model fit, graphical goodness-of-fit diagnostics,
significant tests, confidence intervals, and cost prediction. The lecture will use a real-world example of
an aircraft airframe cost model to help illustrate the methodologies covered.
February 23 Cost Data
J.
Cloos, Institute for Defense Analyses
Emphasizes the requirement and importance
of accurate, consistent and relevant data for use in cost estimating and
analysis. The discussion will focus on
various dimensions of cost data and how such data should be presented for
analytical purposes. The overall cost
data process will be described, to include identifying data requirements,
assessing potential sources, and selecting methods for collecting and adjusting
data. Specific areas to be covered
include definitions of costs and cost structures (e.g., Work Breakdown
Structures), major sources of cost data, both contractor (e.g., Cost
Performance Reports) and government (e.g., Future Years Defense Program),
consistent measurement in dollar terms (e.g., normalization), and the
implications of time value of money on costs and investments (e.g., net present
value).
March 2 Estimating
Relationships - II: Complex Models
P.
Lurie, Institute for Defense Analyses
This lecture is a continuation of the
lecture on “Simple Models.” The lecture
will introduce more realistic non-linear models that are often used to model
cost relationships in practice. The same
aircraft airframe example used in the earlier lecture will be used again to
guide the students through model estimation, diagnostics, testing, and prediction. Also to be covered are methods of cost risk
analysis, which allow the analyst to supplement point estimates of cost with
more informative distribution on
cost.
March 9 Force Costing Handout (3 files)
L.
Roark, Institute for Defense Analyses
Introduces the concepts of force and
infrastructure costing. An example will
be used to demonstrate how the tools and techniques taught in earlier sessions
can be used to create a cost estimate for a change in force structure of a
defined force. Infrastructure categories
will be defined and their relationship to forces will be discussed. Important questions the analysts must
consider will be identified along with possible approaches to overcome inherent
shortcomings of available analytical tools and methods. A brief history of force costing and trends
for the future will be presented.
March 16 Spring Break
March 23 Midterm Examination (Project Proposals Due)
March 30 Cost
Progress Curves Learning Curve Homework
A.
Loerch,
The public generally understands the
notion of the learning curve. This
lecture presents the theory and history of the development of the related cost
progress curve in common use by cost analysts.
The fitting of curve parameters is illustrated using empirical
data. Application of the curve is
illustrated using an example. Extensions
of the cost progress curve are described.
April 6 Production
Rate Effects
S.
Balut, Institute for Defense Analyses
The relationships between direct and
indirect costs, and fixed and variable costs will be discussed, along with
related implications for possible error in forecasts of the costs in future
periods. A method for separating fixed
and variable costs will be presented.
Two methods for forecasting future costs will be described, both of
which take into account the component of cost that is fixed, thereby avoiding
this source of error in estimates.
April 13 Schedule Estimating
B.
Harmon, Institute for Defense Analyses
As proposed defense system acquisition
programs approach major decision points, the reasonableness of planned
acquisition schedules must be assessed by DoD representatives. Research into this problem had included the
development of databases and methods of assessing the reasonableness of
proposed acquisition schedules for tactical aircraft, tactical missiles, and
unmanned space systems. The cornerstone
of the methods consists of time estimating relationships (TER’s). TER’s are analogous to CER’s, where the
dependent variable is a time interval instead of cost. This lecture will focus on both the
development of the methods and their application, illustrated using examples.
J.R.
Nelson, Institute for Defense Analyses
Life-cycle cost estimating relationships
and procedures will be presented in the context of the DoD acquisition process. A hypothetical fighter aircraft will be used
to illustrate the methodology. The
phases of the life-cycle (development, production, operating and support) will
be described, and the components of costs for each element of the fighter
aircraft (airframe, engine, and avionics) will be addressed for these
phases. Constant, then-year and
discounted life-cycle cost funding streams will be discussed.
April 27 Software Costing
T.
Frazier, Institute for Defense Analyses
Describes the background and history of
software cost estimating. The discussion
will focus on software metrics used today; software cost estimation methods;
software cost models; and the predictive accuracy of current models. Finally, the lecture will deal with the
impact of new software engineering practices and tools on software cost
estimating.
May 4 Project
Presentations
A.
Loerch,
Final
Examination Take Home due
Thursday, May 11,
A. Loerch,
Performance
Evaluation:
Midterm Examination 30%
Final Examination 30%
Project in Cost Analysis 40%