Software Multi-project Resource Scheduling: A Comparative Analysis

Traditional scheduling methods from general project management community have high degree of centralization and limited capability to deal with uncertainty, and do not well catered for software projects, and most of them lack scalability to high complexity.

Abstract

Software organizations are always multi-project-oriented, in which situation the traditional project management for individual project is not enough. Related scientific research on multi-project is yet scarce. This paper reports result from a literature review aiming to organize, analyze and make sense out of the dispersed field of multi-project resource scheduling methods. A comparative analysis was conducted according to 6 aspects of application situations: value orientation, centralization, homogeneity, complexity, uncertainty and executive ability. The findings show that, traditional scheduling methods from general project management community have high degree of centralization and limited capability to deal with uncertainty, and do not well catered for software projects. In regard to these aspects agile methods are better, but most of them lack scalability to high complexity. Some methods have balanced competence and special attention should be paid to them. In brief, methods should be chosen according to different situations in practice.