On modeling stochastic dynamic vehicle routing problems

Ulmer, Marlin W.; Goodson, Justin; Mattfeld, Dirk C.; Thomas, Barrett W.

doi:10.1016/j.ejtl.2020.100008

Cited by 83 publications

(21 citation statements)

References 72 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figure 7 indicates that the convergence of CAV of the RL started from step 2,400 for the actor model and step 4,200 for the A3C algorithm. In each training and testing step, these CAV were calculated using Equation (32). Based on Figure 7, it appears that the agent trained by the A3C algorithm yielded the highest CAV for the performed tasks, because the A3C algorithm had more opportunities to explore the transport environment than the conventional actor-critic algorithm.…”

Section: Experimental Results From the Case Study Without Uncertaintymentioning

confidence: 99%

“…Generally, an RL reward function for VRP uses a positive and negative route cost for the reward and penalizes actions accordingly. This type of reward functions was developed in [29], [31], [32]. However, more aspects are required for a thorough evaluation when considering real-world problems not limited only to routing costs.…”

Section: Vrp Models That Combines ML and Rlmentioning

confidence: 99%

“…where i denotes a baseline index, Baseline i denotes a result from the state-of-the-art models, and current solution denotes the result of the proposed model. The RL model was evaluated using the agent action value defined in Equation (32). The optimization were compared with state-of-the-art models using Equation (33).…”

Section: E Performance Evaluation Metricsmentioning

confidence: 99%

See 2 more Smart Citations

A Hybrid Reinforcement Learning-Based Model for the Vehicle Routing Problem in Transportation Logistics

Phiboonbanakit¹,

Horanont

Huynh

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Currently, the number of deliveries handled by transportation logistics is rapidly increasing because of the significant growth of the e-commerce industry, resulting in the need for improved functional vehicle routing measures for logistic companies. The effective management of vehicle routing helps companies reduce operational costs and increases its competitiveness. The vehicle routing problem (VRP) seeks to identify optimal routes for a fleet of vehicles to deliver goods to customers while simultaneously considering changing requirements and uncertainties in the transportation environment. Due to its combinatorial nature and complexity, conventional optimization approaches may not be practical to solve VRP. In this paper, a new optimization model based on reinforcement learning (RL) and a complementary tree-based regression method is proposed. In our proposed model, when the RL agent performs vehicle routing optimization, its state and action are fed into the tree-based regression model to assess whether the current route is feasible according to the given environment, and the response received is used by the RL agent to adjust actions for optimizing the vehicle routing task. The procedure repeats iteratively until the maximum iteration is reached, then the optimal vehicle route is returned and can be utilized to assist in decision making. Multiple logistics agency case studies are conducted to demonstrate the application and practicality of the proposed model. The experimental results indicate that the proposed technique significantly improves profit gains up to 37.63% for logistics agencies compared with the conventional approaches.

show abstract

Section: Experimental Results From the Case Study Without Uncertaintymentioning

confidence: 99%

Section: Vrp Models That Combines ML and Rlmentioning

confidence: 99%

See 1 more Smart Citation

A Hybrid Reinforcement Learning-Based Model for the Vehicle Routing Problem in Transportation Logistics

Phiboonbanakit¹,

Horanont

Huynh

et al. 2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The difference between the dynamic vehicle routing problem with stochastic demands (DVRPSD) with VRPSD lies in the different solving perspective, that is, whether the solving process reflects the dynamic adjustment of vehicle route or not [53][54][55]. [56] studied the route planning problem of multi-stage technicians based on experience service time, [57] proposed a RA framework, [58] studied the VRP with real-time traffic information, and [59] proposed a route-based MDP modeling framework.…”

Section: Othersmentioning

confidence: 99%

The Stochastic and Dynamic Vehicle Routing Problem: A Literature Review

Zhang

Guo³

2021

Proceedings of 2021 Chinese Intelligent Automation Conference

View full text Add to dashboard Cite

Because of its inherent difficulties and wide actual application requirements, uncertain vehicle routing problem has attracted extensive attention recently. Unfortunately, few literature have conducted a comprehensive review on this problem. We give a review mainly from the perspective of stochastic and dynamic, and further classify them into several small categories according to the solution methods and application scenarios respectively. In this survey, we ignore the complex mathematic models and algorithms, aiming to give an intuitive understanding on structure, common methods and strategies for beginner in this field.

show abstract

“…On the other hand, the dynamic stochastic VRP benefits from the problem's prior knowledge. Three strategies, including stochastic modelling [29], sampling method and look-ahead dynamic routing [30] were mainly used in the literature.…”

Section: Related Work and Motivationmentioning

confidence: 99%

Towards Novel Meta-heuristic Algorithms for Dynamic Capacitated Arc Routing Problems

Minku

Menzel

et al. 2020

Parallel Problem Solving From Nature – PPSN XVI

View full text Add to dashboard Cite

The capacitated arc routing problem (CARP) is a challenging combinatorial optimisation problem abstracted from typical real-world applications, like waste collection and mail delivery. However, few studies considered dynamic changes during the vehicles' service, which can make the original schedule infeasible or obsolete. The few existing studies are limited by dynamic scenarios that can suffer single types of dynamic events, and by algorithms that rely on special operators or representations, being unable to benefit from the wealth of contributions provided by the static CARP literature. Here, we provide the first mathematical formulation for dynamic CARP (DCARP) and design a simulation system to execute the CARP solutions and generate DCARP instances with several common dynamic events. We then propose a novel framework able to generalise all existing static CARP optimisation algorithms so that they can cope with DCARP instances. The framework has the option to enhance optimisation performance for DCARP instances based on a restart strategy that makes no use of past history, and a sequence transfer strategy that benefits from past optimisation experience. Empirical studies are conducted on a wide range of DCARP instances. The results highlight the need for tackling dynamic changes and show that the proposed framework significantly improves over existing algorithms.

show abstract

On modeling stochastic dynamic vehicle routing problems

Cited by 83 publications

References 72 publications

A Hybrid Reinforcement Learning-Based Model for the Vehicle Routing Problem in Transportation Logistics

A Hybrid Reinforcement Learning-Based Model for the Vehicle Routing Problem in Transportation Logistics

The Stochastic and Dynamic Vehicle Routing Problem: A Literature Review

Towards Novel Meta-heuristic Algorithms for Dynamic Capacitated Arc Routing Problems

Contact Info

Product

Resources

About