Demand Management

This article discusses how you can avoid artificially inducing instability into your own demand. It identifies the key contributors to instability and begins to suggest some options.

Links to other best practices and training at bottom of page.

There is little point in making a manufacturing process infinitely flexible in order to satisfy demands that are artificially variable. ("Don’t flex the factory. Stabilise the demand", Hal Mather, 1994 APICS Conference.) Our experience suggests that variability is induced in the supply chain by:

Sales processes

Sales and the winning of sales are influenced by the behaviour of sales and marketing organisations. Often these people are motivated with financial incentives by short-term considerations leading to feverish activity spurred by special deals at the end of the accounting period and no sales at the beginning of the following period, which induces lumpy demand. Quantity discounts lead to periods of high activity to produce a large quantity some of which will sit on someone’s shelf until they actually need it. (See Previous Malpractice of the Week 002: BOGOF). Some suggestions to solve this problem are made in "Lean Supply Chains".

Management accounting ("Hitting the Numbers")

Another problem is the monthly and annual accounting cycles where pressure is applied to operations management to ship things to make the numbers at the end of the accounting period. An example of this is shown below for a real but anonymous company that started chasing the numbers at a particular point in the sales graph below (where the units have been removed).

After a steady performance over a period of years they had an initial unacceptable first month’s performance after they became end-of-the-month bean counters, which led to pressure to get a good set of numbers the next month which emptied the supply chain and led to a bad set of numbers the following month etc. The company had degenerated into a cycle of expediting, with totally unreliable demand on it’s manufacturing and supply chain, trying to hit the numbers. Some suggestions to solve this problem are made in "The Future of Manufacturing", but this problem is discussed in detail in Previous Malpractice of the Week 006: Hitting the Numbers (The worst way to run Operations))

Demands communicated to suppliers

The customer frequently does not know what they want until they actually want it, and base their demand statements on forecasts. However the forecast once produced is then forgotten and never measured against reality. The answer of course is to have infinitely flexible resources such that you do not need to forecast. Some suggestions to solve this problem are made in Previous Best Practice of the Week 010: Lead time Reduction Overview.

Interpretation of demand by suppliers

This is a process of second guessing the customers demand, because you have observed that they do not know what they want. In some circumstances this is true and it may by less significant to you which particular product they want but to identify overall demand levels. However there are obvious dangers in this approach, which should be limited to the minimum needed within lead-time. Sophisticated forecasting systems often produce precise but totally inaccurate demand data, because the models on which they are based are invalid for that particular situation. True demand is not only your shipping history but in addition some of your lost sales. Demand is also very difficult to predict if, as a result of not having stock, an order is lost. Also if your customers are stockists one demand may be exaggerated because the end customer visited a number of stockists to try to obtain their item, each of which then interpret this as an unfulfilled demand. There are two solutions to this problem:

• To forecast based not on your immediate customer's demand, but based on the demand on them by their customers. This is discussed further in "Postponement and Mass Customisation", and Reducing forecast error, reducing the need to forecast and avoiding the problems caused by inaccurate forecasts.
• Reduce the need to forecast by reducing your lead-times above.

The planning process

Often MRP systems are used to convert demand into materials and manufacturing requirements, and often these operate in a faulty way because of inappropriate system selection, or faulty system design, implementation, and operation. Faulty MRP systems often give rise to demands that can fluctuate violently within lead-time further up the supply chain. The frequency of planning is also a major consideration and must reflect the dynamism of the business. There is a degree of damping or amplifying inherently involved in planning processes. These must be managed. This is described as "nervousness". Some suggestions to solve this problem are made in "MRP1".

Inventory and production planners

Inventory and production planners frequently distort demand in the planning process. Safety stocks, batch sizes, yield / shrinkage allowances, lead-times are often unnecessarily high to give planners a quite life, or unnecessarily low because of pressures on inventory. Or parameters are unnecessarily dynamic in an attempt to reflect change but which swamp the suppliers' ability to absorb that change, either through their production processes or though their planning processes. A symptom of this is frequent re-planning at the expense of controlling the plan. Much more common however is the inadequate maintenance of planning parameters. Some suggestions to solve this problem are made in "The Future of Manufacturing". The problem of setting batch sizes is discussed in detail in Previous Malpractice of the Week 005: "EOQ / EBQ The worst way to set batch sizes".

Information delays in the process

Delays can occur at any stage and can be temporary or permanent but are always a serious problem. Common causes of delay are sales and purchasing office in-trays, and computer problems.

Physical delays in the process

Physical delays are commonly caused by decision-making delays on quality issues, or unreliable processes, or faulty communication.

___________________________________________________________

The following further best practice articles were also mentioned in this paper:

The following public training courses and in-house workshops provide solutions to managing demand:

SSC01 Tools Techniques & Modern Trends in Supply Chain Management

SSC05 Producing Accurate Forecasts

SSC07 Strategic Supply Chain Management

SSC08 Participative Sales & Operations Planning

To discuss your consulting or training needs with one of our independent consultants or trainers please Contact Us.

 Home Page

Ó SM Thacker & Associates (Consultancy and Training Specialists) Original April 2000 v3 August 2007