SCM Fundamentals: Demand Management

SCM Fundamentals: Demand Management

Production Planner

In recent articles we have spoken about the different types of demand you will see in your business how do you manage all that. Well as we looked at each part, forecasts, customer orders, stock demand, we discussed different methods for managing each one.  Demand management is how we manage all those elements into one process.

At its core demand management is information management.  As we have discussed there are processes for managing the flow of information on forecast, customer orders etc.  Most modern ERP systems will have tools to help you in this process. The likes of SAP, NetSuite, Infor and Sage all contain modules specifically designed to help manage demand.  For companies without a large ERP solution there are still many options in Customer Relationship Management CRM software.  And of course, many small companies still manage demand on excel. In essence it does not matter what system you use as long as you have a comprehensive process for managing demand.  And another important principle to always remember is no matter how good your computer system is, someone must take responsibility for the data. 

Some key overall tips to help your demand management process:

  1. Use standard time parameters. For example, if you want to plan production in weekly buckets look for the forecast in weekly buckets.  This is important because it allows you to easily combine sets of data and eliminates the danger of misinterpretation. 

    Say for example if you are told by one market that they forecast to sell 1000 units in November and another market has already given you firm customer orders for 300 in the first week of November, 200 in the second, 400 in the third, and 300 in the fourth.  How do you combine both of those into one file?  One needs to be changed but which one?  I would always look for more detailed information and opt for the weekly buckets.
  2. Require the same time fences. Depending on the requirements of the business you need to have time fences that suit you and your customers. When do orders become firm and cannot change. How far out does the forecast go? These are elements that are best standardised as much as possible across your customers and markets. There will be exceptions (you may give a large important customer more flexibility than smaller ones) but in general you need to have the lengths of time that rules apply and the lengths of time you want the forecast to look out to be standardised and agreed.
  3. Make sure all the data is in the same base unit. For example, if you supply butter you don’t want one customer requesting 300 cases. A region forecasting in tonnes per month. And a warehouse requesting 1000 units of safety stock.  All these may make perfect sense to the person preparing the data and may be established ways of measuring your product. But if different files have different base units then this will have to be interpreted later.
  4. Make sure all the sources of information use the same file or data structure.  Once you have agreed standards you should make sure the file format is the same.  That way information can be quickly loaded into your database (be that an ERP system or Excel).  In companies that use Electronic Data interchange (EDI) this is not an issue.  When I have done this in the past, I would generally send out excel files in a format that the system needed, and I would require information on forecasts etc back in the same format.  I would also work with my IT department to make sure that I could do an extract of customer orders from the database in the same format.  You would be amazed at the things people can do in excel to mess up your data. I had one upload which repeatedly failed yet a manual check of all the files did not show anything wrong. Eventually I tracked it down to one file. Looking at the data on a screen did not show any particular problems. It was only after some more investigation that we discovered that all the numbers were in text format rather than number format and all had a space before them. So instead of giving a forecast of 10 the file was giving a forecast of “ 10” which caused the import to fail. How do you apply a number to “ “.  So, insist that all data is in a standard format. 

Demand management is a key part of the whole S&OP process but to achieve this it is best if it is regarded as a separate process to the S&OP process because of it’s day to day implications. Many companies will have a Demand Planner who specifically looks after all this information.  They gather the forecast, order information, stock demand etc and feed it to production planning. In large organisations each of those functions will have someone responsible for it.  Of course, this is not always possible depending on the resources available to the company. The demand planning function will often be folded into other functions in smaller companies. However, I would recommend that the function should be handled by someone in the planning department who is not also responsible for a specific element of demand. For example if the person responsible for maintaining safety stock levels is also the demand planner the temptation may be there to give higher importance to the safety stock when there is a supply constraint.