Wednesday, September 24, 2014

Forecasting for Inventory Management

Forecasting for Inventory Management

Blog for Week 4
Yvonne Zhang

Inventory management is important to companies, especially for manufacturers, wholesalers or retailers, which have repetitive orders with suppliers or a new product launched. An optimal solution will be a minimum level of inventory but be able to fill the most of orders from customers, which will increase the inventory turnover ratio, meet orders efficiently and reduce the cost of storage. With this goal, inventory forecasting is a crucial part of operation.

Generally, there are three kinds of method to forecast inventory level. The first one is qualitative method, which relies on the judgment of managers, such as unaided judgment, prediction market, game theory, and simulated interaction. The second one is quantity method, with statistics and calculation, such as extrapolation, casual models and discrete simulation. The last are time series method, end use method, etc.

Basically, there are three elements need to be considered in the forecasting. Lead time, which is the time period between the orders delivered to suppliers and the goods actually received. Stocks are required to be sufficient to meet the demand in this period. Meanwhile, extra stocks are necessary because the goods may not be delivered on time. Besides, seasonal patterns are important in the forecasting. For example, there might be a sales peak at the end of each year.

With the technology development, many tools can help to decide the optimal level of inventory. Take P&G as an example, which is the largest consumer goods company in the world, multiple methods are combined in the forecasting of sales and inventory. Analytics and operation research techniques are implemented. With operation research tools, P&G’s product supply analytics team can effectively decide the best source of product, which saved approximately $67 million each year and reduced the order-and delivery cycle from 20 weeks to 8.

Question is technology is not enough to forecast demand in some cases. It also required the experience and judgment of managers from multiple departments. How could managers in different departments effectively work together to make decisions with high efficiency?


No comments:

Post a Comment

Note: Only a member of this blog may post a comment.