Crashing project time by Linear programming (LP): a case study

Abstract

Business and its environment are more complex today than ever before and decision making has become much more complicated than in the past due to increased decision alternatives, uncertainty, and cost of making errors. As a result, it is very difficult to rely on a trial and error approach in decision making. While dealing with projects, to become competitive, sometimes it is required to complete a project within the predetermined deadline to keep cost at lowest possible level. Failure to do so ultimately leads to increase in total cost. Crashing an activity saves time but increases the cost. Thus there must be a time-cost trade-off. This study mainly provides a framework for reducing total project time at the least total cost by crashing the project network using Linear Programming (LP). Then the model is solved with real project data of an installation project by using Solver in Microsoft Excel. In this work real data is used from machine installation projects in Siemens Bangladesh Healthcare. An illustrative example with a project network consisting of 19 nodes and 22 activities is provided. The computational study includes tabulation of the interrelationships among time and cost. In this work project crashing base on LP real data is used from installation projects in Siemens Bangladesh Healthcare. Siemens Bangladesh Health care deals with different medical equipments in different hospitals of the country. Now Siemens has several projects in hand and to complete those projects in time it is necessary to crash the project time of those projects. One of the projects is AXIOM Artis installation project which is an angiogram machine at Lab Aid Cardiac Haopital. From previous experience it has observed that it took 51 days to complete the full installation. In this project, management decided to crash the project time and set 40 days to complete the installation. The purpose of this study is to implement the linear programming models outperforms the default algorithm implemented in these packages to minimize total project cost. It appears that the programming in project time-cost trade-off concept has more to offer than the graphical methods and scheduling software packages. A real understanding of this concept leads to improved solutions to large scale systems facing project cost control systems and network compression.