A Knowledge and Analytics-Based Framework and Model for Forecasting Program Schedule Performance

This paper discusses a mixed methodological approach to address cost and schedule delays in large-scale engineering programs (LSEP). Research has identified potential causes or factors for schedule delays, such as: ineffective human resources policies and practices, consolidation of the aerospace industry, too many stakeholders, and lack of knowledge-based acquisition practices. Current methods and tools are ineffective in helping project managers to accurately predict schedule performance during LSEP development. The authors describe research to investigate the feasibility of: (1) deriving quantitative and qualitative causal factors correlating to schedule performance during LSEP development; (2) developing a framework and a tailorable predictive model using the causal factors in a Bayesian Network (BN) model; and (3) using the resultant framework and BN model, with expert knowledge elicited from subject matter experts, to predict schedule performance and inform decision makers on actions needed to manage schedule performance. Finally, this paper discusses a version of a BN model developed by mapping a conceptualization of the framework created using a systemigram with a BN pattern that includes dependencies from causal and control factors to schedule performance. The BN incorporates a direct causal dependence of schedule performance and mitigation actions to the consequences of schedule performance.