optimize-project-cost-LP

Optimize project cost given project constraints. Formulate it as constraint optimization problem and solve it.

Problem Description

In project management, we frequently encounter requirements to plan project such that overall project is completed within committed time to customers/users, but with optimized/minimal cost. This is clearly a optimization problem. Below we formulate it as Linear Programming(LP)/Mixed Integer Linear Programming(MILP) problem and resolve it using PuLP, the python LP solver.

Aim

We want optimize/minimize the project cost such that project duration/deadline is within K weeks

Project tasks, their inter dependencies, normal durations, normal costs, crashing durations and crashing costs are given.

Crashing means max extent to which task duration can be decreased, but with increased costs. For eg: A task with normal duration of 1 week with normal cost of 100 dollars, might be decreased to 0.5 weeks(crashing duration) with a (crashing) cost increased to 160 dollars.

Tools used:

1. PuLP (python Linear programming library) with CBC and GLPK backend solvers.
2. Numpy
3. Pandas

Please checkout Jupyter notebook optimize-project-cost.ipynb for details.