[This article was first published in Army Sustainment Professional Bulletin, which was then called Army Logistician, volume 3, number 2 (March–April 1971), pages 12–15, 42–43. The text, including any biographical note, is reproduced as faithfully as possible to enable searchability. To view any images and charts in the article, refer to the issue itself, available on DVIDS and the bulletin’s archives at asu.army.mil/alog/.]
Military spending is a matter of primary concern to the taxpayer today. The pressure is “on” to reduce costs. Significant contributions have been made toward this goal with the development of automated mathematical life cycle cost models.
IN THE PRESENT ATMOSPHERE of heightened interest by Congress and the public-at-large in Defense contracting and costing procedures, it is particularly appropriate to note the concentrated effort that is being made by the military departments to develop more efficient and accurate cost estimating procedures and techniques. Cost analysis developmental efforts, although begun in 1965, were greatly expedited by establishment of the Department of the Army Cost Analysis Program in 1966. Under the direction of the Comptroller of the Army, cost analysis organizations were established at major and subordinate command levels throughout the Department of the Army. Within the U.S. Army Materiel Command, these efforts were formalized last year under PROMAP-70 (Program for the Refinement of the Materiel Acquisition Process in the 70’s). This article reports briefly on the efforts carried out by the Cost Analysis Office at the Headquarters, U.S. Army Weapons Command (USAWECOM), Rock Island, Illinois, before and since the advent of PROMAP-70.
Basic Concept
The functions of the Cost Analysis Office are many and varied, however, its primary effort is directed toward four functions:
• Conduct of cost analysis studies.
• Development of methodology.
• Accumulation and validation of data.
• Review and validation of important command estimates.
Additionally, the office serves as the Cost Estimating Control Center for Headquarters, USAWECOM. The purpose of such a center is to insure a maximum degree of validity and consistency for major cost estimates developed by various elements of the headquarters for outside recipients.
The basic operational concept of the Cost Analysis Office realizes the need for the conduct of a wide variety of cost estimates and studies. These can range from very simple cost estimates requiring only desk calculations to comprehensive life cycle cost studies requiring a high speed computer. These comprehensive studies usually require detailed yearly costs (over a multiyear period) for several mixes of weapon systems containing many different systems per mix.
It has been the experience of this office, however, that in between these two extremes of computational requirements, there is a large variety of studies and analysis work requiring computations too large for hand calculations and yet too small to warrant the cost and time involved in computer operation. This gap is filled very nicely and economically by a rather sophisticated calculator called the Mathatron. This machine operates on punched paper tape and has a rather sizable memory bank. Inputs can also be made on a console, while output is accomplished through a teletypewriter. This machine is particularly useful for simple but repetitive calculations such as regression analysis and the conduct of short, but multiyear, cost studies.
The Mathatron has also made it possible to automate the “debugging” of large computer programs during the development of our comprehensive life cycle cost models. “Debugging” of a computer program refers to the discovery and correction of errors in the computer program (or instructions to the computer). This is generally done by checking the results of computer computations with those done by hand. In the case of some of our computer programs, ranging upwards of some 1,900 statement lines, considerable time is saved by avoiding the necessity for the extensive hand computations involved.
Continuing Requirement
With this background on the basic concept of operation, the remainder of the article will be concerned with the development and evolution of USAWECOM automated life cycle cost models and their use in conducting comprehensive weapon system life cycle cost studies.
The U.S. Army Weapons Command, which supplies the Army with conventional weaponry, such as small arms, artillery, aircraft weapons, fire control, tanks, and other combat vehicles, has a continuing requirement for conducting comprehensive weapon system cost studies. These studies usually involve a determination of detailed yearly life cycle costs for several alternative courses of action. Life cycle costs may be developed for several weapon systems, one weapon system with several quantities and time frames, or a combination of several weapons and many different quantities and time frames per weapon system. Since 1965, the U.S. Army Weapons Command has been developing and employing in its cost studies various automated mathematical models and techniques. All computer programs are written in Fortran language and may be used on any computer capable of using this language.
Small Arms Models
Development of small arms life cycle cost models was begun in 1965 while the Small Arms Weapon System (SAWS) Study was being conducted. The models were developed to meet the requirements for complete and detailed costs for several mixes of small arms weapons. Each of the two SAWS models consists of five or six equations that simultaneously develop life cycle costs for systems (weapons and ammunition) being phased into, as well as those being phased out of, the Army inventory. Both models develop and summarize costs according to the three cost categories of development, investment, and operating costs. One of the two models is more sophisticated in that it develops and summarizes such costs on a yearly basis, whereas the more simplified model develops such costs only on the basis of total costs for a ten- to fifteen-year period.
An important aspect of cost analysis is the problem or how the analyst deals with the uncertainties inherent in the basic information or in the cost factors available to him. This problem really amounts to identifying those cost factors which, to a large extent, control total costs and in determining to what degree specific variations in these important factors will affect the computed total costs. Sometimes, variations in a particularly important cost factor will result in large variations in the final cost estimate.
An overall summary of the models and techniques presented in this article is contained in an unclassified publication that may be obtained from the Defense Documentation Center (AD #688599). This document, “Automated Life Cycle Cost Models for Army Weapon Systems,” published in May 1969 by Headquarters, U.S. Army Weapons Command, presents complete and detailed descriptions of five models for small arms, tank machineguns, and tanks and other combat vehicles, such as self-propelled artillery.
One technique for dealing with this problem is the so-called sensitivity analysis technique. Essentially, this technique consists of conducting the cost study or cost estimate repetitively, each time holding the numerical values of all factors constant except for one, which is varied by a known amount.
An example of the usefulness of cost sensitivity analysis in cost analysis studies is the following illustration of how it was employed to determine the degree of significance of the cost differences existing among three different small arms weapons designated as rifles A, B, and C.
Sensitivity Analysis
Data from the sensitivity analysis study are plotted in figure 1 below to show the impact on total system cost of variation, in each of the four principal cost parameters involved. It is obvious from this plot that variations in Ap (i.e., the average peacetime ammunition cost per weapon per year) cause maximum variations in total system costs. This is true since the cost study assumed peacetime conditions with provisions for normal war reserves.
Figure 2 below is constructed to show how variations in the ammunition parameter alone affect the total cost for each of the three rifles under consideration. In view of the potentially large variation in Ap, resulting from the rather large uncertainties inherent in average ammunition usage rates for these weapons, an analysis of figure 2 leads to the conclusion that there is really no significant cost difference between rifles B and C. This is true since total system cost for rifle B could very easily be less than that for rifle C if the average ammunition cost for C were increased by only 10 percent while that for rifle B were simultaneously decreased by 10 percent from their nominal estimates. However, one could safely conclude that rifle A is likely to cost more than rifles B or C with the maximum difference being nearly two billion dollars over a fifteen-year period in peacetime.
The rather poor confidence in the validity of computed total costs, as revealed in the above analysis, illustrates very forcefully the extreme importance of conducting such analyses. The situation just analyzed is a classical example involving the worst possible situation from a cost uncertainty point of view. This is so since the ammunition parameter represents not only the most important parameter from a cost sensitivity point of view but also the parameter with the largest degree of uncertainty.
Complete Study Made
As a result of this analysis, the decision makers required that the entire study be done for several combinations of estimates of ammunition usage rates for the weapons involved.
Figure 3 on page 42 is a flow diagram that illustrates the basic structure of the combat vehicle model. Each cost element is categorized according to two basic schemes. These are the basic cost categories of development, investment, and operating, represented by the first three columns, and the Army Fiscal Code categories, which are represented as groupings of rows on the chart. Thus, vertical addition of these cost elements represents summarization of costs by the three cost categories, while horizontal addition of these same elements represents summarization according to the Army Fiscal Code categories. The fourth column on the chart represents summarization of costs according to the latter type of categorization. To facilitate comprehension of the basic structure of the model, the detailed cost elements are not shown. However, the chart does indicate the three most important groups of cost elements that are used to compute the principal research and development, procurement and production, and operational and maintenance costs.
Mathematical Model
The mathematical model consists of some 100 equations and computer Fortran statements, requiring approximately 110 sets or elements of input cost factors, schedules, and other data. This model, developed in late 1968 and early 1969, was the most highly developed and flexible of the cost models developed up to that time by the USAWECOM Cost Analysis Office. It is particularly useful in that it has the capability of developing and summarizing total and yearly costs according to the two schemes of categorization illustrated by the chart.
Since its completion in early 1969, this model has been used to conduct several life cycle cost studies involving a variety of weapon systems (i.e., machineguns and artillery as well as combat vehicles). It has permitted the conduct and modification of comprehensive studies, involving several mixes of weapon systems, in a matter of a day or two on very short notice.
By virtue of its capability to handle up to seven mixes of seven weapon systems per mix with one pass through the computer, the model is able to analyze the sensitivity of total costs to seven variations, each one involving seven different cost parameters, with one computer run. The entire routine requires only a matter of seconds on the IBM 360-65 computer.
Pilot ICE Studies
PROMAP-70 has included, during the past year, a series of comprehensive cost studies for important weapon systems. As a part of this series of studies, referred to as Pilot ICE (Improved Cost Estimates), the USAWECOM Cost Analysis Office developed additional life cycle cost models and techniques. These include the development and summarization of life cycle costs according to a third and more extensive and detailed scheme of cost categorization outlined in Department of the Army regulation 37-18. At the present time, in addition to the further development of life cycle cost models mentioned above, that office is experimenting with an entirely different concept for cost modeling. The concept is concerned with development of modular constructed models. According to this concept, a library of cost model subroutines or submodels would be developed. These would be subroutines or limited-restricted models for each phase of the life cycle, for each basic type of weapon system, for various computer output formats (i.e., various ways of presenting the results of cost studies), for various methods of treating the cost uncertainty problem, and for many other purposes. The idea is that eventually, with a varied and well-established library of such subroutines, an analyst will be able to construct readily an automated life cycle cost model, practically tailormade, to apply to any cost problem by a judicious selection of the subroutines. These subroutines could be easily stored on magnetic disc files and easily retrieved in the required order by a few punched IBM cards. This concept was used during the latest Pilot ICE studies on the Bushmaster machinegun and on an artillery piece.
This article has been a brief review of some of the more important developments in life cycle costing within one organizational unit of one subordinate command of the U.S. Army Materiel Command. It is evident that within USAWECOM, as elsewhere throughout the U.S. Army Materiel Command and the Department of Army, the emphasis is on automation to achieve maximum efficiency at minimum cost. One can safely say that the USAWECOM Cost Analysis Office could not accomplish the major portion of its functions, nor respond as rapidly as expected to requests for high quality, comprehensive cost studies with the available resources without automation.
Quality and Validity Essential
The reference to “high quality studies” leads to a second important point that must be emphasized in connection with cost analysis studies in general, and automated models in particular. It is important to realize that the quality and validity of any cost study depends almost entirely upon the quality and validity of the input data, assuming of course, that a rational approach and methodology are used. The development of valid input data is undoubtedly the most important, the most difficult, as well as the most rewarding of all cost analysis tasks. For this reason, the USAWECOM Cost Analysis Office has exerted considerable effort over a two-year period in the development of an automated method to develop average yearly maintenance parts cost factors for our weapon systems. This procedure permits efficient analysis of worldwide parts requisition data maintained on magnetic tapes by the national inventory control point at this command.
In spite of the advances that have been made in recent years, the problems in Army cost analysis are numerous and difficult and there is still much challenging and rewarding work ahead.
Mr. Frederick M. Chakour was chief of the Scientific Systems Division, U.S. Army Management Systems Support Agency, Department of the Army. Formerly an operations research analyst with the U.S. Army Weapons Command, he assisted in developing the Army’s cost analysis instruction. Mr. Chakour held a master’s degree in chemical engineering.
| Date Taken: | 06.26.2025 |
| Date Posted: | 06.26.2025 11:15 |
| Story ID: | 501576 |
| Location: | US |
| Web Views: | 13 |
| Downloads: | 0 |
This work, Automated Life Cycle Costing by Federick M. Chakour, must comply with the restrictions shown on https://www.dvidshub.net/about/copyright.
