Fresh Food Distribution Center Storage Allocation Strategy Analysis Based on Optimized Entry-Item-Quantity-ABC

For labour-intensive field, appropriate storage location assignment is the best choice to increase order picking efficiency and reduce order cycle time, which satisfy customers and reduce cost at the same time. In this paper, we advance a storage location assignment for fresh food distribution center with a manual picker-to-parts picking system by using an optimized approach. To reflect the customer demand uncertainty, the orders received in a certain time range have been grouped and given the different coefficients according to the reference value of them. On that basis, the storage location can be designed optimally based on the Entry-Item-Quantity (EIQ) analysis, which can be used to resolve some orders picking issues, long-picking time and high inventory costs, caused by seasonal change of fresh food, unstable customer demand and repeat purchases. From the computational results, new storage allocation strategy achieves at most a 16% reduction in travel time.


Introduction
Storage location assignment is critical to the warehouse management. Lack of scientific storage policy, any change in demand frequencies or addition to the product assortments leads to high costs-the very problems posed by fresh e-commerce enterprises.
Many scholars have conducted in-depth research about assigning products to storage location, resulting in a large number of research results. These researches are mainly focused on 3C products, tobacco, grain oil and so on. Compared with these products which requirements are static on the whole, fresh food requirements are with the characteristic of seasonal. Conventional methods of analyzing logistics centers mainly focus on examining data onto lump sums or single-period orders instead of multiple-period orders. Consequently, such methods are not appropriate for managing rapid changes in contemporary market orders.
Through the comprehensive analysis of e-commerce storage assignment policy, this paper makes an innovative research methods addressing fresh food seasonal demand. The proposed method is divided into two stages. In the first stage, the orders in a certain time range have been grouped and given the different coefficients according to the reference value of them. In the second stage, the storage location can be designed optimally based on EIQ-ABC method. Figure 1 shows the proposed procedure with the optimized EIQ-ABC for classifying inventory items as A, B or C and rearranging them accordingly.

Problem Formulation
The fresh food distribution center consists of storage zone (freezer), production zone and buffer zone, used to keep fresh meat, eggs, aquatic products, etc. Freezer is the most labour-intensive place of fresh food distribution center workforce. Some medium and small-sized fresh food distribution center usually needs to deal with 1000 orders a day and the number will expand to about 2000 orders during the holidays or promotions days. Order picking is the most important parts of a distribution center, which is the "heart" of the distribution center". There are usually some problems existed in the fresh food distribution center.
(1) The order picking efficiency is low. Among the warehousing activities in distribution centers, order picking is the most time-consuming and labour-intensive. Warehousing activities including picking, processing, sorting, packaging and distribution take 2h at least. Order picking (picking by article) generally accounts for 64%. As a result, order picking becomes a bottleneck preventing distribution centers from maximizing the effectiveness of their warehousing activities. There's the research data we got by tracking the warehouse activities, as shown in table 1 and table 2. (2) There's no reasonably effective storage assignment in freezer. Under the random-storage policy, pickers get accustomed to putting the goods in some empty storage location arbitrarily, which increases the difficulty of searching goods; any addition to the products leads cargo to occupy channels.
(3) There are mostly outdated equipped and inefficient picking system in warehouse, which makes the pickers spend a long time to get the order goods. Order picking error rates increase significantly with the growth of consumption demand.
Freezer storage location design has become important because of its impact on service to customers and total logistics costs. The performance of freezer deeply affects the competency of fresh food distribution center.

Analysis of Influence Factors and Principles
The best location assignment scheme is usually satisfied the operational requirements that it supports. Du Zizheng believes that frozen products storage is based on the shelves system, which reasonable design and layout not only can be convenient to the operation, but also takes full advantage of storage space. It is very necessary to analyze the influencing factors. We take three factors into account: (1) The quality of fresh food makes greater demands on the logistics technology and management. Their main target customer groups are the generation after 80 and 90, the so-called "net born generation". Besides enjoying the convenience of online shopping, they pay greater attention to the experience of online shopping and goods quality.
(2) Fresh food is highly sensitive to the logistics cost and effectiveness. Perishability requires more rapid delivery, so as to improve the satisfaction degree of the target customer groups.
(3) Fresh food includes four principal features: big quantity of order, small quantity of a single order, high order frequency and high repurchase rates. The storage layout should reflect the demand characteristics of products, which means we should consider some decision variables, such as the frequency of goods delivery, the correlation of demand and so on.
At the same time, some principles should also be considered when storage equipment, sorting tools and sorting strategy in certain circumstances: (1) Matching Job requirements principle. The storage location of goods is built on its demand frequency. The higher ones are stored closer to the I/O point and the lower ones are located farther away from the I/O point, by which the response time of the system can be reduced.
(2) Matching product features principle. By storing fresh food according to the correlation of customers demand, pickers can quickly search for the goods in demand.
(3) Be mindful of "outside line" principle. Generally speaking, there are one or two kinds of goods with low frequency in each order. If you worried about traveling through the "outside line" to fetch goods, it should be setting the special storage position according to distribution centers situation.

Data Processing
Order data from late January to mid-February is selected from WMS. The order data are divided into three groups (1.15-1.23, 1.24-1.31, 2.1-2.7) and set weights. For the latest order data can reflect the customer's consumption better, group one (1.15-1.23) weighted coefficient is 0.2, group two (1.24-1.31) weighted coefficient is 0.3 and group three (2.1-2.7) weighted coefficient is 0.5.
Partial screenshots of the processing order data are shown as table 3, table 4 and table 5: Grand Uruguay refinement boneless meat 400g 10 5 Hormel export record 500g (Beijing) frozen lean meat stuffing 3 1.5

EIQ-ABC Analysis
According to the EIQ analysis demand, the weighted processed data is sorted out: (1) IQ analysis Make IQ analysis table and partial screenshots of the table, as shown in table 6: Sincere the products name is too long, we set up different code-name for fresh food respectively to facilitate the statistics. Pareto Analysis chart is shown as figure 2: As can be seen from the graph 2, A-items products account for only 20%, while its outgoing cumulative percent reach 80%; B-items account for 15% and its outgoing cumulative percent is 10%; C-items products are a small proportion of outgoing cumulative percent, while the percentage of product up to 63%.
(2) IK analysis Make IK analysis table and partial screenshots of the table,  as shown in table 7  Then, make Pareto analysis chart, as shown in Figure 3: As can be seen from the graph 3, A-items products account for only 10.71%, while its order cumulative percent reach 50.14%; B-items account for 20.24% and its order cumulative percent is 27.86%; C-items products are a small proportion of order cumulative percent, while the percentage of product up to 69.05%. The difference is not so significant in goods order times. So during analyzing how to classify the freezer fresh food, we mainly tend to the results of IQ analysis.
Combining the analysis above, items are clustered into three classes in such a way that 93 items are called A-items, 61 items are called B-items and 263 items are called C-items. Locations closer to the depot are used to store A-items, aiming at shorter traveling distance and less handling workload. The storage locations are arranged according to travel distance to the I/O point in an ascending order: Location A should be the largest storage area, using dedicated storage. In order to cope with the explosive growth of orders, inventories can be increased to a certain extent; In the Location B, the storage area should be large, which can employ class-based storage; The shipments of the fresh food in Location C is not high, random storage should be used. C-items are assigned to a small number of shelves located in a small area or in the fixed area to improve the freezer space utilization (figure 4).

Conclusion
At present, random storage assignment is adopted in the freezer. And there's no fixed picking route, pickers find goods according to their preference for picking and familiarity with the warehouse.
Under the storage strategy the paper proposed, orders need to be generated A-items batches and other items batches separately, which are picked by different pickers at the same time.  The table 8 shows that with the new storage policy, a significant reduction in travel time can be achieved. Operation efficiency of the whole logistics center has been optimized with the new freezer storage location strategy, which also could reduce increasing pressure of fresh food distribution centers.
Today, order picking is more difficult because distribution centers have to store more types of products to fulfill more small orders within a shorter response time. To be more effective means corporations have stronger competitive ability.