Multidimensional project control system
Project control systems often fail to support management in achieving their global project goals. This thesis proposes a Multidimensional Project Control System (MPCS) as an approach for quantifying deviations from the planning phase to the execution phase with respect to the global project control specification (GPCS). The projects' current state must be translated into yield terms, which are expressed as a gap vector that represents the multidimensional deviation from the global project control specification. The MPCS methodology allows the project manager to determine: integrated project status; where problems exist in the project; when and where to take corrective action; and how to measure improvement. However, implementing the MPCS methodology does not require extra data collation. MPCS deals with the control of a single project and defines the project performances in comparison with the plan. The progression of several projects in parallel is a common situation in organizations, therefore a comparison of the various project performances is required. It is proposed that a comparison process be performed using the data envelope analysis (DEA) approach. The reference points for examining the performances of different projects and the directions of improvement for the projects are not necessarily found on the efficiency frontier. An algorithm is developed for applying multi-project system control having a relatively large number of inputs and outputs while maintaining the validity of the DEA methodology. The DEA output allows the diagnosis of those found on the efficiency frontier and those that need improvement.