Adaptable Process Model Checklists: Cost Estimates and Schedule
	IMPORTANT NOTICE: The complete Adaptable Process Model (APM) is provided for informational purposes and for assessment by potential users. The APM is copyrighted material and may not be downloaded, copied, or extracted for use in actual project work. The full hypertext (html) version of the APM may be licensed for use and customization within your organization. Contact R.S. Pressman & Associates, Inc. for complete licensing information.
	Cost Estimates and Schedule The intent of this checklist is to assess the "reality" of project cost estimates and schedule (SEPA, 5/e, Chapters 3, 4, and 7). It should be noted that experience may be the best arbiter when evaluating the Software Project Plan. However, the following issues should be addressed by those examining the plan estimates and schedule. For this checklist, the more questions that elicit a negative response, the higher the risk that the estimate is inaccurate. Cost Estimates Is the application/system to be built similar in technology and function to other applications/systems that have been built by the project team? Was some formal mechanism (e.g., FAST, use-cases) used to elicit basic requirements before the estimation task was undertaken? Is the project scope bounded? Have estimates of project size (e.g., LOC or FP) been developed? Have size estimates been generated by more than one person? Do the people making the estimates have experience in the application domain? Do the people making the estimates have at least three years of application experience? Was decomposition of function used to generate size estimates? Was the functional decomposition taken to at least two levels of detail? Are all functions (that were refined) understood technically? Has a feasibility analysis been performed for all major functions? Has the impact of reusable software components been considered? Has a detailed consideration of make vs. buy been conducted? Has a decision tree been created? Is it likely that major application/system functions will remain unchanged? Is it likely that secondary application/system functions will remain unchanged? Has a reasonable (mathematical) method been used to reconcile different size estimates? Does a historical database containing project metrics for past software projects exist? Are reliable averages for $/LOC and/or $/FP available for the application domain to be addressed by this project? Was a software process model chosen before the estimates were made? Were framework activities and work tasks selected before estimates were made? Were effort estimates attached to all framework activities? Were effort estimates attached to all work tasks? Has a reasonable (mathematical) method been used to reconcile differences between size estimates and independent effort estimates? If one or more empirical cost models or cost estimation tools have been used, have they been calibrated for the local environment? Project schedule Would the majority of project team members consider the project deadline as achievable? reasonable? Would the customer be willing to accept late delivery if the delay were caused by conditions beyond the project team’s control? Is a project scheduling tool (e.g., MS Project) being used to manage the schedule? Has a complete work breakdown structure for the at least the first 20% of chronological schedule been developed? Have intertask dependencies been clearly identified? Does the preliminary allocation of project effort across the project schedule approximate the 40-20-40 rule? Have measurable milestones been developed at points no more than 5% apart on the project timeline? Are project milestones attached to SQA activities such as FTRS? Are project milestones attached to the production of reviewable software engineering work products? Have individual team members been assigned to work tasks? Have timeline charts been generated for each team member? for the project as a whole? Do the team members assigned to software engineering tasks have appropriate training and experience for those tasks? Has the project manager/technical leader managed projects of this size and duration before? Are human resources scheduled at no more than 80 percent of their time? Has the project critical path been computed? Is the project critical path assessed regularly for any changes? Are critical project tasks independent of activities performed by resources outside the control of the project team? Is each work task tracked for budgeted vs. actual completion date? Have contingency plans been developed to handle delays in critical tasks? Is earned value analysis being used to assess progress? Have error tracking mechanisms been established throughout the project? Return to Checklist Table of Contents