Beyond the Edge of Feasibility: Analysis of Bottlenecks

Abstract

The productivity of real-world systems is often limited by so-called bottlenecks. Hence, usually companies are not only interested in finding the best ways to schedule their current resources so that their benefits are maximized (optimization), but, in order to increase the productivity, they also conduct some analysis to find bottlenecks in their system and eliminate them in the most efficient way (e.g., with the lowest investment). We show that the current frequently used analysis (based on average shadow price) for identifying bottlenecks has some limitations: (1) it is limited to linear constraints, (2) it does not consider all potential sources for bottlenecks in a system, and (3) it does not provide adequate tools for decision makers to find the best way of investment to eliminate bottlenecks and maximize the profit they can gain. We propose a more comprehensive definition of bottlenecks that covers these limitations. Based on this new definition, we propose a multi-objective model for the benefit and investment. The solution for this model provides the best way of investment in resources to achieve maximum profit. As the proposed model is multi-objective and non-linear, it opens an important opportunity for the application of evolutionary algorithms, which can subsequently have a significant impact on the decision making process of companies.