ABSTRACT
The project compression problem is a subproblem of the planning problem. Typically, the project level control (work method) is held fixed and only lower level controls are optimised. A number of approaches have been suggested in the literature for solving
the project compression problem as an optimisation problem, typically a linear optimisation or linear programming problem. The optimisation approaches do not appear to have found favour in commercial project planning programs or with practitioners. Some reasons suggested for this are:
• Most commercial project planning packages appear to be based on algorithms that reflect the particular characteristics of networks. In this way the algorithm can be written in general terms and applied to all networks. A linear programming approach, on the other hand, requires input peculiar to each network – the network logic of activity precedence is interpreted as constraints in the linear programming approach. It is conceivable that a transformation could be written in general terms giving the network logic as constraints.
