A network representation procedure for the optimization of capital budgeting problems is proposed. Interrelated capacity expansion projects and discrete time decisions are considered. An alternative-based approach is proposed that yields a very neat formulation by identifying the relationships among alternatives before solving the problem. The most general pairwise interactions are considered, and a 0-1 integer program is formulated with nonlinear objective function and linear constraints. Procedures are demonstrated that convert the nonlinear program into a linear form with an embedded network structure. The out-of-kilter algorithm is applied to solve the resulting network flow problem. Because of the inherent nature of the network structure, this approach enables efficient calculation of optimal solutions for relatively large problems. Computational efficiency is shown that requires less than 3 min of central processing unit time for a system with 372 alternatives.
|Title of host publication||Transportation Research Record|
|Number of pages||9|
|Publication status||Published - 1993 Dec 1|
All Science Journal Classification (ASJC) codes
- Civil and Structural Engineering
- Mechanical Engineering