Implementation of the critical chain project management methodology in IBM's S/390 software development environment

Thesis (S.M.)--Massachusetts Institute of Technology, System Design and Management Program, February 1999."December 1998."Includes bibliographical references (p. 91-92).Software Development projects have a long history of being notoriously difficult to manage. From early experiences with the IBM OS/360 Operating System over 20 years ago to more recent experiences with the IBM OS/390 Operating System, the Project Management challenges remain. This phenomenon exists despite the wide spread availability of well-developed Project Management techniques such as the Critical Path Method (CPM) and the Project Evaluation and Review Technique (PERT). The challenges also transcend the introduction of ever more powerful programming tools and techniques such as structured programming, high-level languages, source-level debuggers, and object-oriented programming. As in many industries, a key challenge in the Software industry is the reliable delivery of products in an environment of ever decreasing product cycle times. Recent work by Eliyahu M. Goldratt suggests that the struggle with on-time delivery may well lie with the underlying Project Management techniques that have become so widely accepted. These techniques foster behavior patterns that are counter-productive to the shortening of product cycle times. They fail to focus the organization on the Project Management system at large and can encourage dysfunctional decision making [l ]. Work in the field of System Dynamics has independently reached similar conclusions. The traditional Project Management techniques offer little to help the Project Manager cope with issues at the strategic level. Without strategic guidance, the Project Manager is left to make poor, informal judgments and may not make adequate allowances for factors that negatively impact project performance [2]. Goldratt offers a new, alternative project scheduling approach called Critical Chain as a mechanism for improving an organization's underlying Project Management structure. Critical Chain is based on principles developed a decade earlier in Goldratt's Theory of Constraints. The Theory of Constraints changed the way organizations think about Manufacturing processes. Likewise, Critical Chain requires that organizations reformulate their approach to managing development projects. This thesis will study the successful results of applying Critical Chain on two actual Software Development projects in IBM's System 390 Division. Each of these projects achieved commitments on time. Critical Chain's contribution to these results will be discussed. The experiences gained along with potential pitfalls of Critical Chain will also be considered. In particular, the issues involved with applying this approach to a Software Development environment in which traditional methods are in widespread use will be emphasized. A discussion of the potential limitations of the Critical Chain approach will also be provided.by Theodore R. Maeurer.S.M