Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Hybrid Memetic Search for Electric Vehicle Routing with Time Windows, Simultaneous Pickup-Delivery, and Partial Recharges

Zubin Zheng, Shengcai Liu, Yew-Soon Ong

TL;DR

This work tackles EVRP-TW-SPD by introducing a Hybrid Memetic Algorithm (HMA) that combines large-neighborhood search with population-based memetic search. The two key innovations are Parallel-Sequential Station Insertion (PSSI) for handling partial recharges and Cross-Domain Neighborhood Search (CDNS) to exploit both EVRP-TW-SPD and VRP-TW-SPD solution spaces. A new large-scale benchmark derived from JD Logistics data in Beijing is proposed to better reflect real-world tasks, alongside an open-source implementation. Across small, medium, and large instances, HMA consistently outperforms state-of-the-art Adapt-CMSA variants, with notable gains on large-scale problems thanks to the BCD decomposition and the cross-domain search framework. The study demonstrates the practical relevance of advanced hybrid heuristics for realistic EV routing under time windows and partial recharges, with implications for scalable, eco-friendly logistics planning.

Abstract

With growing environmental concerns, electric vehicles for logistics have gained significant attention within the computational intelligence community in recent years. This work addresses an emerging and significant extension of the electric vehicle routing problem (EVRP), namely EVRP with time windows, simultaneous pickup-delivery, and partial recharges (EVRP-TW-SPD), which has widespread real-world applications. We propose a hybrid memetic algorithm (HMA) for solving EVRP-TW-SPD. HMA incorporates two novel components: a parallel-sequential station insertion (PSSI) procedure for handling partial recharges that can better avoid local optima compared to purely sequential insertion, and a cross-domain neighborhood search (CDNS) that explores solution spaces of both electric and non-electric problem domains simultaneously. These components can also be easily applied to various EVRP variants. To bridge the gap between existing benchmarks and real-world scenarios, we introduce a new, large-scale EVRP-TW-SPD benchmark set derived from real-world applications, containing instances with many more customers and charging stations than existing benchmark instances. Extensive experiments demonstrate the significant performance advantages of HMA over existing algorithms across a wide range of problem instances. Both the benchmark set and HMA are to be made open-source to facilitate further research in this area.

Hybrid Memetic Search for Electric Vehicle Routing with Time Windows, Simultaneous Pickup-Delivery, and Partial Recharges

TL;DR

This work tackles EVRP-TW-SPD by introducing a Hybrid Memetic Algorithm (HMA) that combines large-neighborhood search with population-based memetic search. The two key innovations are Parallel-Sequential Station Insertion (PSSI) for handling partial recharges and Cross-Domain Neighborhood Search (CDNS) to exploit both EVRP-TW-SPD and VRP-TW-SPD solution spaces. A new large-scale benchmark derived from JD Logistics data in Beijing is proposed to better reflect real-world tasks, alongside an open-source implementation. Across small, medium, and large instances, HMA consistently outperforms state-of-the-art Adapt-CMSA variants, with notable gains on large-scale problems thanks to the BCD decomposition and the cross-domain search framework. The study demonstrates the practical relevance of advanced hybrid heuristics for realistic EV routing under time windows and partial recharges, with implications for scalable, eco-friendly logistics planning.

Abstract

With growing environmental concerns, electric vehicles for logistics have gained significant attention within the computational intelligence community in recent years. This work addresses an emerging and significant extension of the electric vehicle routing problem (EVRP), namely EVRP with time windows, simultaneous pickup-delivery, and partial recharges (EVRP-TW-SPD), which has widespread real-world applications. We propose a hybrid memetic algorithm (HMA) for solving EVRP-TW-SPD. HMA incorporates two novel components: a parallel-sequential station insertion (PSSI) procedure for handling partial recharges that can better avoid local optima compared to purely sequential insertion, and a cross-domain neighborhood search (CDNS) that explores solution spaces of both electric and non-electric problem domains simultaneously. These components can also be easily applied to various EVRP variants. To bridge the gap between existing benchmarks and real-world scenarios, we introduce a new, large-scale EVRP-TW-SPD benchmark set derived from real-world applications, containing instances with many more customers and charging stations than existing benchmark instances. Extensive experiments demonstrate the significant performance advantages of HMA over existing algorithms across a wide range of problem instances. Both the benchmark set and HMA are to be made open-source to facilitate further research in this area.

Paper Structure

This paper contains 25 sections, 1 theorem, 10 equations, 3 figures, 6 tables, 3 algorithms.

Table of Contents

  1. Introduction
  2. Related Work
  3. Notations and Problem Definition
  4. Hybrid Memetic Algorithm for EVRP-TW-SPD
  5. Parallel-Sequential Station Insertion (PSSI)
  6. Station Selection
  7. Charge Amount Calculation
  8. Discussion
  9. Cross-Domain Neighborhood Search (CDNS)
  10. Aggressive Local Search (ALS)
  11. Conservative Local Search (CLS)
  12. Discussion
  13. Overall Framework of HMA
  14. Experiments
  15. Experimental Setup
  16. ...and 10 more sections

Key Result

Proposition 1

Given an EVRP-TW-SPD instance, removing all charging stations from a feasible EVRP-TW-SPD solution always results in a feasible VRP-TW-SPD solution.

Figures (3)

  • Figure 1: The flow chart of HMA.
  • Figure 2: Critical difference (CD) plots across various problem instances using the Nemenyi post hoc test.
  • Figure 3: Comparative effects of PSI and SSI on each subset. The left stacked bar illustrates the exclusive contribution ratio (ECR) of SSI and PSI, with the remaining portion corresponding to shared contributions. The right stacked bar depicts the time allocation ratio (TAR) of SSI and PSI, with the remainder representing the TAR of other algorithm components.

Theorems & Definitions (1)

  • Proposition 1