Research On The Optimization Of Storage Location Assignment And Order-picking Routing In A Shelf-type Warehouse

In recent years, with the development of global business, logistics has gradually become the focus of some enterprises. Many logistics enterprises, and even large-scale retail enterprises and electronic business companies are building their own warehouse centers and distribution system so as to improve the service level and reduce logistics cost. In warehouse centers, order picking time is the 40% of logistics cost. So, order-picking operation efficiency directly affects the efficiency of the whole warehouse center and operation cost. The reasonable storage location assignment is the precondition of efficient picking operations. The reasonable order-picking routing is an effective measure to reduce the picking personnel walking time and labor. Therefore optimization methods of storage location assignment and order-picking routing are studied for warehouse center with shelves in this thesis.Firstly, the storage location assignment problem is studied and a clustering-assignment method is developed. Through analysis on the correlation of historical order items with the historical order frequency, a clustering-assignment model is built, and a heuristic algorithm is presented, which is implemented by MATLAB software. Finally, a case is applied to analyze the results and compared with popular frequency-based assignment storage method. It is validated that the total picking distance is reduced by 2.35% with the clustering method than the popular frequency-based assignment storage method, and the parking times is reduced by 16.67%.Secondly, the picking routing problem is formulated. According to the different characteristic of the warehouse centers, divide the picking routing problem into single-stock-location problem and multiple-stock-location problem. Then it is further divided into One-Order-One-Truck problem and One-Order-Many-Truck problem. In this thesis, each combined problem has been optimally solved by using the genetic algorithm. The Multiple-Stock-Location and One-Order-Many-Truck problem, the most complex one, is studied and solved based on a case, considering load factor and the location deviation of items to be picked. Finally, the effect of pickup capacity, location deviation of items and item quantity (the number of orders and item quantities are include) in orders on the total picking distance is analyzed.