Time: Tuesday, June 25, 2019, 10:30-11:30
Venue: A1716 Report Hall, Science Building, Zhongbei Campus
Topic: Large-scale planning warehousing optimization based on Markov Monte Carlo simulation
Reporter: Feng Xingdong, Professor, Shanghai University of Finance and Economics
Abstract
Warehousing planning refers to the decision-making and design of warehousing mode, facilities and information management system before warehousing activities. The problem of warehousing planning studied in this paper is to optimize the location of materials in warehouse under given warehouse storage rules and technology conditions, aiming at maximizing economic benefits. Starting from the actual storage problem of automobile companies, we use Gibbs sampling to deal with complex and high-dimensional sampling problems, and improve the simulated stochastic approximation annealing algorithm to solve the multi-level and multi-constrained combinatorial programming problem. The validity of the proposed optimization algorithm is verified by calculating the real storage data of automobile companies. Compared with the simulated stochastic approximation annealing algorithm, the Markov Monte Carlo simulation algorithm for warehouse planning designed by us has advantages in computing accuracy and solving time.
Brief Introduction of the Reporter:
Feng Xingdong, Permanent Professor and Doctor of Statistics and Management College of Shanghai University of Finance and Economics, Executive Director of the Institute of Big Data Research. Ph.D. at the University of Illinois, Champaign, and postdoctoral at the National Statistical Center of the United States. He has been engaged in the scientific research of econometrics, mathematical statistics and biological statistics for a long time, and has published more than 30 papers in four high-level journals of statistics. Presided over projects of self-Science funds in many countries.