A5079619929- Vehicle Routing Optimization Methods
- Scheduling and Optimization Algorithms
- Elevator Systems and Control
- Transportation and Mobility Innovations
- Smart Parking Systems Research
- Advanced Manufacturing and Logistics Optimization
- Autonomous Vehicle Technology and Safety
- Electric Vehicles and Infrastructure
- Assembly Line Balancing Optimization
- Vehicular Ad Hoc Networks (VANETs)
- Advanced Graph Neural Networks
- Advanced Neural Network Applications
Waseda University
2022-2024
Job shop scheduling problem (JSSP) is one of the well‐known NP‐hard combinatorial optimization problems (COPs) that aims to optimize sequential assignment finite machines a set jobs while adhering specified constraints. Conventional solution approaches which include heuristic dispatching rules and evolutionary algorithms has been largely in use solve JSSPs. Recently, reinforcement learning (RL) gained popularity for delivering better quality In this research, we propose an end‐to‐end deep...
Vehicle routing problem (VRP) is one of the classic combinatorial optimization problems where an optimal tour to visit customers required with a minimum total cost in presence some constraints. Recently, VRP being solved use deep reinforcement learning (DRL), node sets considered (represented) as graph structure. Existing Transformer based DRL solutions for rely only on information ignoring role edges between nodes In this paper, we proposed attention‐based end‐to‐end model solve which...
Abstract The job shop scheduling problem (JSSP) is a well-known NP-hard combinatorial optimization that focuses on assigning tasks to limited resources while adhering certain constraints. Currently, deep reinforcement learning (DRL)-based solutions are being widely used solve the JSSP by defining structure disjunctive graphs. Some of proposed approaches attempt leverage structural information capture dynamics environment without considering time dependency within JSSP. However, graph...
The capacitated vehicle routing problem (CVRP), which is referred as NP-hard a variant of Traveling Salesman Problem (TSP). CVRP constructs the route with lowest cost without violating capacity constraints to meet demands customer nodes. Following advent artificial intelligence and deep learning, use reinforcement learning (DRL) solve giving promising results. In this paper we proposed DRL model CVRP. transformer-based encoder our fuses node edge information construct rich graph embedding....
The performance of a neural network relies on the depth model to learn structural correlations features. Nevertheless, Graph Neural Networks (GNN) tends lose its efficiency as increases. In this paper we propose technique alleviate problem by building existing GNN architecture. effect, installed gating mechanism overcome propagation noise information across layers and trained using proximal policy optimization (PPO), gradient-based reinforcement learning algorithm. We proposed capacitated...