Content area
The U.S. inland waterway transportation system (IWTS) is an essential part of the country’s multimodal freight transportation network. In addition to seasonal droughts and floods, the operations of the IWTS may be disrupted by random malfunctions and scheduled maintenance of its critical components. Among these critical components, locks play a key role in the operation of navigable inland waterways, and lock-induced disruptions to the supply chains of related industries, such as agriculture and manufacturing, often result in significant economic losses. To assess the performance of the U.S. IWTS, we develop a PyNetLogo simulation tool to capture the movements and delays of cargoes considering various sources of uncertainty such as water level, lockage time, and lock failure. Using this simulation tool, a series of lock repair and preventive maintenance actions are determined via Deep Reinforcement Learning (DRL) to minimize the loss due to lock-induced disruptions. To illustrate the proposed modeling and decision-making method, the McClellan-Kerr Arkansas River Navigation System is considered in our case study, where a random policy and a first-come, first-served policy conventionally implemented in practice are also presented for comparison. The results show that the optimal strategy obtained by the proposed DRL-based approach outperforms the conventionally implemented alternatives in various aspects. Most importantly, the levels of availability of all the locks are significantly improved, enabling a more seamless cargo flow along these navigable inland waterways. For the benefit of stakeholders, our further study reveals that employing multiple full-time repair crews instead of one can further increase the availability of the locks, but the idle time of these maintenance crews becomes more significant. This provides a way of thinking about the recruitment, deployment, and utilization of maintenance crews responsible for the smooth operation of such critical infrastructures.