Life-Cycle Cost Estimation


Definition: Cost analysis is "the process of collecting and analyzing historical data and applying quantitative models, techniques, tools, and databases to predict the future cost of an item, product, program, or task." Cost estimates "translate system/functional requirements associated with programs, projects, proposals, or processes into budget requirements, and determine and communicate a realistic view of the likely cost outcome, which can form the basis of the plan for executing the work [1]."

Keywords: budget, cost analysis, cost benefit, cost estimation, regression analysis, tradeoffs

MITRE SE Roles and Expectations: Systems engineers (SEs) are expected to use cost analysis to identify and quantify risks and to evaluate competing systems/initiatives, proposals, and tradeoffs. They are expected to collaborate with the cost/benefit analyst and the sponsor to define the approach, scope, products, key parameters, and tradeoffs of the analysis. SEs support and provide direction to the analyst, review results, guide and evaluate the sensitivity of the analysis, and provide technical, programmatic, and enterprise-wide perspectives and context for the analyst.

Cost analysis is often misunderstood and frequently overlooked. The practice encompasses many areas of a program's business management and combines the knowledge of many different disciplines. And, the results of cost analysis have far-reaching impacts on a program and its success. In many cases, the analyst who built a program's life-cycle cost estimate (LCCE) will have more knowledge and understanding of the program than any other member of the program team.

Cost Estimate Development Overview

Cost estimation methodologies and techniques vary widely depending on the customer and program. These variations are based on several factors. What is being estimated, the extent of available data, existence of an agreed-on work breakdown structure (WBS), regulatory requirements, agency requirements, and industry best practices all influence the methodologies and techniques that may be applied when creating a cost estimate. For example, the LCCE for a Department of Defense weapon system will be conducted differently and look very different from an estimate for a data center or the development of a computer application for a civilian agency. Also, the type of estimate influences the methodology and approach used. A much more rigorous process is required for a budgetary estimate or a full LCCE than for a rough order of magnitude or "back of the envelope" type of estimate.

The GAO Cost Estimation Guidebook [2] depicts a detailed, thorough methodology for completing a life-cycle cost estimate. Not every estimate requires the rigor of an LCCE; there is no "cookie-cutter" approach to developing a cost estimate. Figure 1 depicts a generic cost estimating process that aligns with the GAO process, with fewer steps. The specific process an estimator chooses will be directly linked to the level of rigor the type of estimate requires, i.e., higher rigor for an LCCE, less so for a rough order of magnitude (ROM).

Figure 1. General Depiction of Cost Estimating Process
  1. Define the Cost Estimate Scope: The initial step is to define the possible scope of the cost model and the purpose for conducting the estimate. The scope determines the content of the cost elements that must be included in the model. Sources for scope definition of a program include the project management plan, the scope statement, the WBS, and any requirements documentation, etc. The result of this step should be a plan for conducting the estimate. The SE and the cost estimator should work together to define the proposed build in terms of the resources required to create it. The cost analyst needs more than a general statement of desired capability or functionality. He or she needs to know how the engineers are going to build it, i.e., what is the full technical solution? 
  2. Identify Assumptions and Constraints: Assumptions are statements that are used to limit the scope of the model. They are "givens" as opposed to "facts." They usually relate to a future occurrence and therefore contain uncertainty. Assumptions must be evaluated during sensitivity analysis. Constraints are usually fixed, externally imposed boundaries such as schedule, policies, and physical limitations.
  3. Develop Cost Element Structure: The cost element structure can also be thought of as a chart of accounts. It lists the possible categories of cost contained in the model. Each element must be defined so that all costs are covered and no costs are duplicated within the structure.
  4. Collect and Normalize Data: Cost data is collected for all the elements within the model. Information from benchmark research and actual cost experience is used. The normalization process ensures that cost data are comparable.
  5. Develop Cost Estimating Relationships: Cost data is used to develop equations that will be entered into the cost model. The equations are the basis for estimating costs as a function of system capacity and service level. The result of this step should be a point estimate of cost.
  6. Document the Approach: Document each cost element, indicating the sources of data, assumptions used, and any equations used in the calculations. Also document any risk or sensitivity analysis that was conducted.
  7. Conduct a Customer Review: Walk the sponsor through the model and results to ensure that all expected costs have been adequately represented and to achieve acceptance of the estimate.
  8. Finalize the Cost Estimate: Update the cost estimate to reflect actual costs as they are incurred and changes that occur as the program matures.

Best Practices and Lessons Learned

The three R's. For an LCCE to be credible and effective, it must meet three basic requirements, also known as the three R's of cost estimation: Replication, Rationale, and Risk.

  • Replication: The estimator must provide a sufficiently detailed audit trail, including clearly stated assumptions for each cost element, that would allow a third or external party to independently replicate the estimate.
  • Rationale: The estimator must provide a convincing and justifiable rationale for the selection of key parameter values, labor estimates, cost factors, assumptions, and all underlying inputs to the estimate. These can come from early project experience, other similar projects, parametric models, and documented engineering judgments..
  • Risk: The estimator must conduct risk/sensitivity analysis to assess the impact of the inherent uncertainty in input values. Regression analysis is the most frequently used method of conducting sensitivity analysis in this area.

Utility to the program. Investments require clear identification of benefits, which can be either tangible or intangible. The benefits can be packaged in a program that will most likely yield desirable outcomes. As a cautionary step, consider the cost to stand up the program, the cost to incur the chain of activities for the identified investment such as implementation, operation, and maintenance from both quantitative and qualitative perspectives. When properly done, cost analysis provides the following utility to the program:

  • Supports budgeting process by:
    • Integrating the requirements and budgeting processes.
    • Assessing affordability and reasonableness of program budgets.
    • Providing basis for defending budgets to oversight organizations.
    • Quickly/accurately determining impacts of budget cuts on program baselines and associated functionality.
  • Enables early identification of potential pitfalls such as cost growth and schedule slips.
  • Enables identification of future cost improvement initiatives.
  • Provides for the identification and objective quantification of the impact of program risks (technical and schedule risks).
  • Provides a basis for evaluating competing systems/initiatives (cost/benefit analyses and analysis of alternatives [AoA]).
  • Enables proposal pricing and evaluation of proposals for cost reasonableness (independent government cost estimates).
  • Captures cost impacts of design decisions to facilitate tradeoffs in cost as an independent variable/design to cost/target costing.
  • Facilitates evaluation of the impact of new ways of doing business (e.g., in-sourcing vs. outsourcing, commercial off-the-shelf vs. custom software).

An art, not a science. As with any discipline, the actual application and practice of cost analysis is more difficult than the academic description. It is seldom the case that the process outlined above can be applied with complete precision. In most cases, many factors conspire to force the SE and the cost estimator to step "outside the box" in completing a cost estimate.

When data is unavailable. Data to support the estimate is often not readily available through the customer organization. Finding supportable data often requires creative thinking and problem solving on the part of the SE and cost estimator. An example is an AoA in which one of the alternatives was to build roads. The agency in question did not possess any in-house knowledge on road construction, ancillary costs (such as drainage ditches and easements), or permit and legal requirements for the construction of several hundred miles of access road. The situation required reaching out to the civil engineering community and several state departments of transportation in order to bridge the knowledge gap and obtain the information in question. This resulted in a detailed and supportable estimate that the customer was able to use in justifying managerial decisions.

Adaptability is key. As stated in the cost estimation development discussion, there is no single way to construct a cost estimate—too much depends on the details of the circumstances at hand. An estimator cannot do a parametric estimate, for example, if the data and situation do not support that approach. Another AoA provides an example of this. When tasked to provide an AoA for an outsourcing or internal development of a sponsor's financial management system, the estimator predetermined that an engineering build-up based on engineering knowledge of the problem set would be performed. Unfortunately, the sponsor organization had no internal engineering expertise in this area. The estimator was forced to change the approach and build an estimate based on an analogy of similar systems and industry benchmark studies.

Keep program needs in sight. Overall, the most important perspective on cost estimating is to keep the process in context. Remember that the cost estimate is not an end in itself, but rather a means to an end. Understand what the program needs to accomplish through cost estimation and work with the cost estimator to tailor your product accordingly.

References and Resources

  1. The International Society of Parametric Analysts and the Society of Cost Estimating and Analysis (ISPA/SCEA), Professional Development and Training Workshop Proceedings, June 2–5, 2009, training presentation on Cost Estimating Basics [slide 5].
  2. Government Accountability Office, GAO-09-3SP, March 2009, GAO Cost Estimating and Assessment Guide: Best Practices for Developing and Managing Capital Program Costs.

Additional References and Resources

, accessed November 25, 2015.

, accessed November 25, 2015.

, accessed November 25, 2015.

, accessed November 25, 2015.

, Carnegie Mellon University, accessed November 25, 2015.

Publications

Download the SEG

MITRE's Systems Engineering Guide

Download for EPUB
Download for Amazon Kindle
Download a PDF

Questions?
Contact the SEG Team