Economical Optimization of Reversible Lane Systems - A study of modeling and optimizing the economic aspects of reversible lane systems in Gothenburg

Abstract

Reversible lane systems are a strategy which aims to maximize the traffic flow by switching the direction of one, or several, lanes at a road segment. While the implementation of reversible lane systems has been studied in terms of optimizing the total travel time in a traffic system, the economic impacts has historically often been overlooked. In this report a mathematical model is developed that optimizes the economic output of implementing reversible lane systems. The suggested model is a bi-level optimization model. The upper level maximize the economic output through altering the lane configuration and the lower level minimize the total system travel time in accordance with the methodology for System Optimum. The model is evaluated through applying it on central Gothenburg, Sweden. Here a Genetic Algorithm, provided in MATLAB’s Global Optimization Toolbox, is used to implement the model. Due to the restrictions imposed by this algorithm the bi-level programming model is implemented as two separate optimization problems. The two models are used iteratively where the output of the former is used as input in the latter, and vice versa, until an optimized lane configuration is generated. In the end it is concluded that it would not be profitable to implement reversible lane systems in Gothenburg with the current traffic level. Moreover, it is concluded that while the model can be used to generate an indication of the potential profitability of reversible lane systems more developments are needed before basing large infrastructural investments on results generated by this model.