6 An Innovative Distributed Project Control
- 1. -Super Lightweight Tank- A Risk Management Case Study in Mass Reduction AN INOVATIVE DISTRIBUTED PROJECT CONTROL APPROACH : A CASE STUDY OF THE CASSINI MIS Presented by Yulhendri Saputra & Marcel Bernadi
- 2. Introduction Problem Issues Environmental Scanning Alternatives Course of Actions Key Success Factors Selection of Best Alternatives Recommendation
- 3. INTRODUCTION Originally. Congress approved funding for a joint program composed of a Comet Rendezvous/Asteroid Flyby (CRAF) mission and the Cassini mission. The program was authorized with the stipulation that it would be canceled if original cost estimates were exceeded. In response, Jet Propulsion Laboratory (JPL) chose to implement a classical earned-value system for controlling schedule and cost. In 1990, after purchasing a full-featured commercialoff-the-shelf (COTS) project management system, the project hired 20 schedulers and planners to implement a full) integrated network schedule. The system was based on classic program evaluation and review technique (PERT) concepts (developed in 1958) and the Cost/Schedule Control System Criteria (C/SCSC) formalized by the U.S. Department of Defense in 1967. A large expenditure was made for training at all project levels. By late 199 1 , CRAF/Cassini project managers observed somt: trends with the entire project control process: This refereed c a s e study article w a s accepted January 1996.
- 4. ENVIRONMENTAL SCANNING The COTS project management system process countered JPL's effort to having decision making shifted to lower levels in the organization. Considering the large expenditure for project control, no corresponding improvement was made with regard to communication between the different project elements. The COTS project management system was not viewed as an easy-to-use tool. The COTS project management system did not provide a straightforward method for addressing uncertainty.
- 5. PROBLEM ISSUES "Network-based techniques as planning tools do not consider the issue of uncertainty in identifying tasks and the relationship among tasks in the network" (Morad and Vorster, 1993). "A network models only one set of schedule options at any given time" (Mathews, 1993). "The current approach to planning and controlling complex projects has traditionally been to eliminate mcertainty and change from our projects, or at ieast treat them as if change will not occur7' (Archibald and Lichtenburg, 1994). "Older packages emphasize the critical path method, which can be very clumsy when scheduling people. Classic approaches tend to be expensive and roo!ed in legacy technologies" (Wood, 1995). In short, the project found that the cleanliness of theory (PERT, CISCSC) was no match for the clutter of reality.
- 6. ALTERNATIVE COURSE OF ACTION 1. CPM and PERT Network-based techniques as planning tools do not consider the issue of uncertainty in identifying tasks and the relationship among tasks in the network" Eliminate uncertainty and change from our projects, or treat them as if no change Older packages emphasize the critical path method, which can be very clumsy. In short, the project found that the cleanliness of theory was no match for the clutter of reality. 2. CMlS • All project products have to be someone's deliverable (Del) and another's receivable • Once a deliverer and receiver reach an agreement on the need t • Each Reciving and Delivery agreement • Any person on the project is able to determine a Rec/Dells status at any given time
- 7. KEY SUCCES FACTOR The value of this approach became immediately apparent: I. The CMIS schedule control module facilitated and increased communication. It also highlighted where communication between project elements was deficient. Product specification and delivery schedule disputes were easily identified, problems were quickly addressed. and schedule confusion vanished. 2. Schedule tracking and statusing became virtually effortless. The project implemented an automated statusing system that polled receivers on a weekly basis about whether they had received their agreed-to deliveries for that week. Receiver responses to the messages were automatically compiled and summarized in weekly metric reports to management. In addition, on-line statusing within CMIS was provided. 3. The single-database design allowed current project-level schedule reports to be generated at any time. 4. CMIS reduced centralized project control costs to less than 0.25% of the total project budget. The workforce of planners/schedulers dropped from 20 people to an operations staff of 3 . This yielded a savings of greater than 60 work- years over the project's development phase.
- 8. SELECTION OF BEST ALTERNATIVES CMIS Schedule Control Module Planning Reports and Metrics. •Before CMIS implementation, many seasoned managers expected that the simple level of formality (that is, requiring technical managers to input only the product's name, source, destination, and delivery date) would not be beneficial. This perception changed dramatically when only 3 of the initial 1,200 products submitted to the CMIS database had matching receiver and deliverer information. Moreover, technical managers quickly recognized that CMIS allowed them to formalize their external delivery commitments without limiting their ability to modify their internal schedules to best meet their delivery commitments. •The reconciliation process found holes in the plan where needed products were not being supplied or were being supplied but not needed. The process also uncovered many occurrences of product delivery dates that were much later than those needed by receivers. •CMIS provides metrics for tracking the level of reconciliation at the project. system. and subsystem levels. • RecIDel metrics are reviewed weekly by the project manager and his staff. Lack of a reconciled plan immediately draws management attention.
- 9. Recommendation • For large research and development projects, effective cost and schedule control is possible without incurring the large costs associated with centralized project control systems. • An effective cost/schedule management information system for all levels of management can be built by focusing on satisfying the requirements of first-line managers and lead engineers. Such an approach removes the inherent noise in management reports created by planning and control staffs. • An effective way to deal with schedule uncertainty is to simplify the project control approach to focus on the areas where schedule uncertainty has the greatest impacts (that is, the interfaces between receivers and deliverers).
- 10. Thank you Ref : Tuman, John, "It's Time to Re-engineer Project Management." PM Network, 7: 1 (January 1993), pp. 40-4 1 . Morad, Ayman A., and Michael C. Vorster, "Network-based Versus A 1-Based Techniques in Project Planning." Project Management Journal, 24: 1 (March 1993), p. 25. Mathews, Mark D., "Introducing Networks to an In-Progress Project." Project Management Journal, 24:2 (June 1993), p. 9. Archibald, Russell D., and Lichtenburg Steen, "Uncertainty and Change Require a New Management Model." PAM NETwork, 7:5 (May 1994), pp. 6-10. Wood, Lamont, "Project Management Perfect Harmony.“ nformation Week,No. 526 (May 8, 1995), pp. 42-54.
- 11. Thank you 1.Please elaborate the different between PERT project evaluation review technique VS CPM critical path method ,CMIS contstruction management information system ? 2.What are the benefit and disadvantages of using PERT/CPM/CMIS ?