About

Wei Yuan

Research interests: Supply Chain Optimization, Production Scheduling Optimization

Industrial Engineering | 09/2018 - 06/2021

Average Score: 83/100
Major Courses: Data Structure and Algorithm Analysis, Machine Learning, Operations Research, Engineering Matrix Theory
Master's thesis: Task Scheduling and Path Planning in Mobile Robot Fulfillment Systems

Industrial Engineering | 09/2014 - 06/2018

GPA: 3.49/4.0, Outstanding Undergraduate Graduates (5%)
Major Courses: Operations Research, Advanced Mathematics, Programming & Algorithmic Language

Algorithm Engineer (Optimization) | 06/2022 - 07/2024

Combined with historical passenger flow data, established a 0-1 integer programming model
Applied Genetic Algorithm to generate bus schedules that adapt to various types of vehicles in Java

Considering constraints such as order interval and vehicle capacity, an integer programming model was established
Implemented Adaptive Large Neighborhood Search in Java to realize real-time allocation of orders, and completed simulation report

Considering constraints such as target output and operating time, and aiming to minimize the total transportation cost, an integer programming model was established
Utilized Java to develop a Simulated Annealing algorithm, including various neighborhood operators such as adding/subtracting trucks, adding/subtracting tasks, etc., and introduced an operator adaptive selection mechanism to realize static task allocation of mining trucks

Executed A* algorithm in Java to achieve the global path planning of unmanned trucks in automated container terminal
Based on congestion information, realized real-time route re-planning for trucks to avoid congested roads
Implemented a multi-vehicle collaboration algorithm at intersections to avoid conflicts or deadlocks
Applied Monte Carlo Tree Search to determine the order in which trucks pass at intersections
Developed a simulation system in Python, and utilized OpenCV to implement dynamic demonstration and simulation of the system operation process

Algorithm Engineer (Optimization) | 07/2021 - 04/2022

Completed scene function and efficiency tests of path planning program, and wrote a test report
Constructed the automatic online program in C++ for AMRs in warehouse which can bring malfunctioning robots back to the topological map automatically
Modified C++ code to adapt the path planning program to omnidirectional vehicles, debugged and simulated on the server to achieve stable operation of 20 AMRs
Investigated related papers on path planning sequence optimization of AMRs, and completed a research report

Software Development Intern | 07/2020 - 09/2020

Learned basic knowledge of OpenGL rendering pipeline and shader, and achieved basic particle effects using C++

Implemented A* algorithm and Q-Learning in Python to generate global path for single robot
Applied Deep Q Network to realize multi-robot collision and deadlock avoidance using Python

Established an integer programming model which considered power constraint of robots
Utilized CPLEX to get exact solutions for the task assignment problem in small-scale
Designed heuristic rules and applied Ant Colony Optimization algorithm in C++ to solve the large-scale calculation cases efficiently

Applied Genetic Algorithm in C++ to intelligently adjust the vehicle sequence in the automobile assembly workshop
Designed a local search algorithm to achieve resequencing of mixed-model assembly lines using C++
Won the Academy's Outstanding Undergraduate Graduation Project (10/200)

W. Yuan and H. Sun, ‘A Task Scheduling Problem in Mobile Robot Fulfillment Systems’, in 2020 12th International Conference on Advanced Computational Intelligence (ICACI), Dali, China: IEEE, Aug. 2020, pp. 391–396. doi: 10.1109/ICACI49185.2020.9177514. [Paper]
H. Sun and W. Yuan, ‘Multi-AGV motion planning based on deep reinforcement learning’, Computer Integrated Manufacturing Systems, vol. 30, no. 2, pp. 708–716, 2024, doi: 10.13196/j.cims.2021.0607. [Paper] (Written in Chinese)

Last Updated: Dec. 19, 2024