On Convexity of Optimal Multi-Commodity Freeway Network Control

TL;DR

This paper extends convex relaxation techniques to multi-commodity freeway network control, enabling efficient optimization of ramp metering and variable speed limits for different vehicle classes, demonstrated through a California case study.

Contribution

It introduces a convex formulation for multi-commodity FNC problems, allowing for efficient optimal control with commodity-specific variables, advancing previous single-commodity approaches.

Findings

Convex relaxation is effective for multi-commodity FNC.

Commodity-specific controls outperform single-commodity models.

Case study confirms practical applicability of the method.

Abstract

We study a multi-commodity Freeway Network Control (FNC) problem aiming at achieving optimal operation of a transportation network through the use of ramp metering and variable speed limits. Straightforward formulations of both single- and multi-commodity FNC problems based on the Cell Transmission Model are known to be non-convex, mainly due to the congestion effects at diverge junctions. However, recent studies have shown that it is possible to formulate a tight convex relaxation of the single-commodity FNC problem. We extend these results to the multi-commodity FNC problem by considering concave commodity-specific demand functions and concave aggregate supply functions, so that different variable speed limits can be applied to different commodities. Hence, it is possible to efficiently compute the optimal control action to reduce congestion phenomena in the network. We also present a case study of a segment of the freeway network in California, using data from the PeMS database, to demonstrate the effectiveness of the proposed solution. Finally, we draw a comparison with a setting where the multi-commodity flows are modeled and controlled as a single-commodity flow, to emphasize the relevance of acting separately on different classes of vehicles.