SCM 7.0x - Demand Planning (DP) forecasting using KISS

[jlbeaudin]

Start with the business strategy and keep it simple

The motivation behind my new blog is based on many years of experience of business planning and the design and development of tools supporting multiple customer environments and business models.  My team and I support SCM every day and are constantly challenged with peers and customers taking the wrong approach to SCM.  Most people configure first and then define requirements as they go.  SCM is a complex tool that can have a significant impact on an organizations bottom line – which is why it deserves a clear requirements, business process approach where the people who configure SCM need to first dive deep in the business operations. In this blog I try to provide an introduction to those readers who may have SCM and want to learn more about it. Or to those readers who have SCM and may be unhappy with its performance today.

I’m always open to feedback – please use the comments section to continue the discussion or message me directly from my SCN profile.

• Definition: Forecasting is a process to create a statement of projected demand for a product/customer over time

• What are some Goals for SCM Demand Planning?
  1. Provide long term critical information to the business
  2. Enable the company to accurately plan and optimally manage its resources
  3. Provide the company with on-going cost savings in manufacturing, warehousing, purchased materials, product breakage and spoilage
  4. Optimize customer satisfaction against supply chain costs

What is Demand Planning & Accurate Forecasting?

• A toolkit of statistical forecasting techniques
• Tightly linked to the R/3 System and the SAP BW SAP (data can be automatically transferred)
• Tree selection and drill-down capabilities facilitates navigation through multidimensional data structures
• Uses the Alert Monitor to report exceptions

In utilizing SAP’s Demand Planning the tendency is to work on improving forecasting methods to improve the forecast accuracy and rely on improving the information utilized in generating the forecast. However, focusing on the business requirements and strategy should be considered first, before working toward forecasting accuracy.

As an example if you have a product which has a profit before taxes (PBT) of 20% and another product which has 10% PBT, would you give the same weighting (focus or energy for accuracy) to these two products? Also you may have a customer or group of customers you want to receive a higher level of service than other customers.

Note: products and customers are just two dimensions and businesses may have to consider multiple dimensions to be successful.

What’s required? Is an analysis of your information and the determination of how best to apply the business strategy to these results. If the information states that 80% of your highest profit items is 10% of your products, then possibly you’d put more emphasis on these products, than the other 90% of products. This analysis should be repeated on the Customer dimension and possibly on other critical dimensions (manufacturing facilities, distribution centers, etc…) based on your business. In many cases the business will have to consider two or more dimensions in this analysis and meld these into one overall strategy which enables the success of the business.

An example of this melded strategy would be as follows:

• Service level one: Top Ten key customers (all products), based on Sellable Available Market (SAM) and Total Available Market (TAM) – equals 80% of the total sales
  • Goal: 95% accuracy to the forecast
  • Vendor Managed Inventory (VMI), Kanban or Just In Time warehousing (JIT)
  • Build to Stock or Build to Forecast strategies

• Service level two: Next Thirty customers and key product types, based on growth potential and penetration capability
  • Goal: 80% accuracy to the forecast
  • Distribution centers with Direct ship or Drop ship
  • Build to Stock or Build to Order strategies

• Service level three: Distribution accounts
  • Goal: 80% accuracy to the forecast
  • Build to Stock or Build to Order strategies

• General Service level: Probably more related to spot buys, make to order strategy, new business
  • No specific forecasting goal
  • Build to Order strategy

Following the design of the business strategy the statistical forecasting methods and planners knowledge tools can be utilized to determine the best solution to anticipate future demand. Outlining your market opportunities and manufacturing capabilities will help determine what forecasting methods to use. When determining a forecast method, start at a business level (Business Unit) and manage exceptions (product or product/customer) as necessary and review these exceptions frequently.

Forecasting accuracy in many cases is more of an art than a science, you may achieve 80 percent accuracy with the statistical methods, but the other 15 percent will be the knowledge added to this process.