A study of the planning and integrated cyclic analysis of serial system operations
This research presents a new methodology for the planning and scheduling of construction projects. The methodology is called PICASSO, an acronym for Project and Integrated Cyclic Analysis of Serial System Operations. It is centered on the modeling of resource and activity interactions, and is suitable for cyclic as well as acyclic projects.^ PICASSO integrates modeling elements from traditional scheduling methodologies, mainly from the Critical Path Method, a scheduling technique traditionally used for acyclic projects, and CYCLONE, a simulation methodology mainly used for cyclic construction process analysis.^ The system uses a network of connected rectangles and circles to model activities and resources, respectively. Repetitive activity chains can also specified and performed a number of times with a minimum of modeling effort.^ Scheduling algorithms consist of a Forward Pass and a Backward Pass, similar in objectives to the procedures of the same name in the Critical Path Method, but implemented using simulation algorithms. Results include activity float information as well as resource utilization.^ The system is implemented using several software packages controlled by a core program. Projects are developed on an interactive package based on the Critical Path Method, while part of the algorithms are performed using a Simulation package. Part of the processing is performed by programs developed for this investigation. The nature of the scheduling information required the use of a combined chart developed for this research.^ The research includes Conclusions emphasizing the need for an integrated and resource centered approach to construction planning. Possible future research directions are also pointed out. Several appendices include the source code of programs developed for the system, and scheduling examples. ^
Major Professor: Daniel W. Halpin, Purdue University.
Off-Campus Purdue Users:
To access this dissertation, please log in to our