Neural Network Based Large Neighborhood Search Algorithm for Ride Hailing Services

Syed, Arslan Ali; Akhnoukh, Karim; Kaltenhaeuser, Bernd; Bogenberger, Klaus

doi:10.1007/978-3-030-30241-2_49

Cited by 12 publications

(6 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The results outperform [9,44] for VRP and SDVRP for different node sizes and show competitive results with LKH3, unified hybrid genetic search (UHGS). Also, [75] proposes using [9] model combined with Large Neighborhood Search (LNS) to solve VRPWT in ride-hailing services. They train the neural networks using supervised learning for the insertion operation inside LNS.…”

Section: A Learning Methods For Facilitating Non-learning Methodsmentioning

confidence: 99%

Learning to Solve Vehicle Routing Problems: A Survey

Bogyrbayeva¹,

Meraliyev²,

Mustakhov³

et al. 2022

Preprint

View full text Add to dashboard Cite

This paper provides a systematic overview of machine learning methods applied to solve NP-hard Vehicle Routing Problems (VRPs). Recently, there has been a great interest from both machine learning and operations research communities to solve VRPs either by pure learning methods or by combining them with the traditional hand-crafted heuristics. We present the taxonomy of the studies for learning paradigms, solution structures, underlying models, and algorithms. We present in detail the results of the state-of-the-art methods demonstrating their competitiveness with the traditional methods. The paper outlines the future research directions to incorporate learning-based solutions to overcome the challenges of modern transportation systems.

show abstract

Section: A Learning Methods For Facilitating Non-learning Methodsmentioning

confidence: 99%

Learning to Solve Vehicle Routing Problems: A Survey

Bogyrbayeva¹,

Meraliyev²,

Mustakhov³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…For the reconstruction phase, the authors developed a mixed-integer linear program as a repair method. Lastly, Syed et al [23] proposed a neural network in an LNS setting to solve a vehicle ride-hailing problem. However, it uses supervised learning, which requires a large training dataset and, furthermore, can only perform as well as the algorithm used for its generation.…”

Section: Literature Reviewmentioning

confidence: 99%

GRAPH Reinforcement Learning for Operator Selection in the ALNS Metaheuristic

Johnn

Darvariu

Handl

et al. 2023

Communications in Computer and Information Science

View full text Add to dashboard Cite

ALNS is a popular metaheuristic with renowned efficiency in solving combinatorial optimisation problems. However, despite 16 years of intensive research into ALNS, whether the embedded adaptive layer can efficiently select operators to improve the incumbent remains an open question. In this work, we formulate the choice of operators as a Markov Decision Process, and propose a practical approach based on Deep Reinforcement Learning and Graph Neural Networks. The results show that our proposed method achieves better performance than the classic ALNS adaptive layer due to the choice of operator being conditioned on the current solution. We also discuss important considerations such as the size of the operator portfolio and the impact of the choice of operator scales. Notably, our approach can also save significant time and labour costs for handcrafting problem-specific operator portfolios.

show abstract

“…Most studies focus on the optimization of operational parameters such as the fleet size or the user cost structure [1,[4][5][6] or on efficient algorithms to the Vehicle Routing Problem [7,8]. Another, relatively smaller, focus lies on the integration with public transport [9][10][11][12].…”

Section: Introductionmentioning

confidence: 99%

Addressing spatial service provision equity for pooled ride‐hailing services through rebalancing

Schlenther

Leich

Maciejewski

et al. 2022

IET Intelligent Trans Sys

View full text Add to dashboard Cite

Ride-hailing services have the potential to play a major role within future transport systems. While the operation as well as environmental, social and economic consequences are research subjects, the focus in the literature lies on system efficiency and profitability. In this study, vehicle relocation strategies are investigated with respect to service provision equity using an agent-based simulation approach. The aim is to identify ways of operation that support social fairness while maintaining the profitability and effectiveness of the service. The results show that a typical demand-anticipatory relocation strategy leads to an imbalance in the spatial distribution of mean wait times for the use case of Berlin. This can be significantly improved by applying different relocation strategies. As a consequence, the overall demand for the system rises while the ratio of empty vehicle kilometres increases slightly.

show abstract

Neural Network Based Large Neighborhood Search Algorithm for Ride Hailing Services

Cited by 12 publications

References 13 publications

Learning to Solve Vehicle Routing Problems: A Survey

Learning to Solve Vehicle Routing Problems: A Survey

GRAPH Reinforcement Learning for Operator Selection in the ALNS Metaheuristic

Addressing spatial service provision equity for pooled ride‐hailing services through rebalancing

Contact Info

Product

Resources

About