Related Experiment Video
Updated: Jun 10, 2025

Virtual Prism Adaptation Therapy: Protocol for Validation in Healthy Adults
Published on: February 12, 2020
Parallel adaptive large neighborhood search based on spark to solve VRPTW
Songzuo Liu1,2, Jian Sun3, Xiaohong Duan1
1Faculty of Information Engineering, Shandong Huayu University of Technology, Dezhou, 253034, China.
A new Spark-based parallel Adaptive Large Neighborhood Search (Spark-ALNS) algorithm effectively solves the vehicle routing problem with time windows (VRPTW). This advanced method finds optimal solutions faster and more accurately than existing algorithms.
Area of Science:
- Operations Research
- Computer Science
- Artificial Intelligence
Background:
- The vehicle routing problem with time windows (VRPTW) is a complex combinatorial optimization challenge.
- Existing algorithms often struggle with finding optimal solutions efficiently for large-scale VRPTW instances.
Purpose of the Study:
- To develop a novel, efficient, and accurate algorithm for solving the multi-objective VRPTW.
- To leverage parallel computing with Adaptive Large Neighborhood Search (ALNS) for improved performance.
Main Methods:
- A Spark-based parallel Adaptive Large Neighborhood Search (Spark-ALNS) algorithm was designed.
- Key components include a novel simulated annealing cooling strategy, CW initialization, and tailored destruction/repair operators.
- A new parallel strategy was implemented to enhance accuracy and reduce computation time.
Main Results:
- The Spark-ALNS algorithm achieved superior performance on Solomon's VRPTW instances.
- It found known optimal solutions for 41 out of 56 instances and discovered new optimal solutions for 31.
- Runtime was 3-5 times faster than other parallel algorithms.
Conclusions:
- The proposed Spark-ALNS is a highly effective method for solving the VRPTW.
- It demonstrates significant improvements in solution quality and computational efficiency compared to existing evolutionary algorithms.
Related Concept Videos
Cluster Sampling Method
To choose a cluster sample, divide the population into clusters (groups) and then randomly select some of the clusters. All the members from these clusters are in the cluster sample. For example, if you randomly sample four departments from your...
Parallel Processing
Prismatic Beams: Problem Solving
The design begins with analyzing the beam as a free body to identify moments and force balances, thereby determining support reactions. Next, the...
Area Computation by the Alternative Coordinate Method
Random Sampling Method
Distributed Loads: Problem Solving

