Approximate dynamic programming for automated vacuum waste collection systems

The collection and treatment of waste poses a major challenge to modern urban planning, particularly to smart cities. To cope with this problem, a cost-effective alternative to conventional methods is the use of Automated Vacuum Waste Collection (AVWC) systems, using air suction on a closed network of underground pipes to transport waste from the drop off points scattered throughout the city to a central collection point. This paper describes and empirically evaluates a novel approach to defining daily operation plans for AVWC systems to improve quality of service, and reduce energy consumption, which represents about 60% of the total operation cost. We model a daily AVWC operation as a Markov decision process, and use Approximate Dynamic Programming techniques (ADP) to obtain optimal operation plans. The experiments, comparing our approach with the current approach implemented in some realworld AVWC systems, show that ADP techniques significantly improve the quality of AVWC operation plans. © 2015 Elsevier Ltd. All rights reserved.