Facing equity in transportation Network Design Problem: A flexible constraints based model

- A model of Equity based Network Design Problem is proposed.
- Equity concept is quantified.
- A new horizontal equity performance indicator has been implemented.
- Three different optimizations have been performed in a sensitivity analysis.
- An application to a real size network validates the results.

In transportation planning, solutions designed to meet objectives of equity and social inclusion have to be achieved. From this standpoint, most of Network Design Problem (NDP) models aim at identifying the optimal layout of transportation networks by deterministic bi-level problems formulation to reflect the different goals of at least two decision makers (the network users and the planner).Considering the societal function of transport systems, the NDP, even at operational level, should be addressed also to the research of solutions in which the outcomes of the design (costs and/or benefits) are distributed as much as possible among the potential users or classes of users.Traditional approaches often neglect equity goals that, conversely, should play an important role or define some flows rebalancing problem. In this paper an Equity Based NDP is proposed, where the optimal layout of a road network is determined by minimizing the total system cost under flexible constraints, (jointly with rigid thresholds) by solving a single programming problem. It considers both horizontal and vertical equity criteria in the form of an equity constraint specified for uncertain variables or approximate reasoning environment: this results in a multi-objective fuzzy programming model that aims at maximizing user satisfaction according to all constraints, while taking into account the route choice behaviour of network users.A sensitivity analysis has been performed on a test network, and a comparison with a literature equity based model has been carried out; then, the proposed method has been applied to a real sized network, where an application of equity network design optimization is presented and the results discussed.