The BASF Group’s Intermediates division develops, produces and markets a comprehensive portfolio of about 700 products for coatings, plastics, pharmaceuticals, textiles, and more. Crucial to the success of this operation is Demand Planning and Forecasting. We have a best in class Demand Planning organization that sets the industry standard for best practices. Here I’ll reveal our process, and the small changes we have made that driven significant results.

The Demand Planning Process at BASF

Our Demand Planning starts with robust statistics that are generated with state of the art models in APO Demand Planning and R. Our statistics are of a high enough quality that we typically don’t have to modify them before inputting them. In cases where statistics must be modified to develop an effective forecast, they still serve as a viable baseline reflecting historic sales and trends. Statistics also serve as a time saver, replacing many manual efforts.

Once this is done, Marketing and Supply Chain hold Validation meetings, where forecast discussions take place, looking at every perspective, including collection of additional data from other stakeholders, forecast methodologies vs. accuracy, forecast value add, error, bias, effect on inventory, effect on delivery metrics and continuous improvement.

The process is also examined strategically, with process improvements arising from lessons learned over the years, best practices and state of the art thinking. Evaluation is quantitative and part of the discussion. Forecast accuracy and cumulative error are mainly used, but other metrics are used as well, sometimes on an ad hoc basis to address difficulties.

Our division uses one set of numbers, with Marketing committing in terms of a financial forecast, with Supply Chain’s using the same numbers for production, supply, operative inventory and delivery.

The Unique Forecasting Challenges At BASF

The main forecasting challenge we face is our position in the Value Chain. We are further up the Value Chain from Retail and the Consumer. Many of our Customers are also intermediaries in the process of manufacturing consumer goods. This makes us subject to the activities of our Customers, such as inventory increases/reductions, minimum and economic order quantity adjustments and financial strategy changes. At our position in the Value Chain, these can affect sales as much as Consumer behavior. We are also sensitive to the Bullwhip effect, with slight Retail changes causing much greater changes on our end. All of these are major considerations for our forecasts.

Even though process-wise we have met many challenges by streamlining and carefully analyzing data, POS data and multi-tiered analysis are less valuable to us than they are for businesses further down the Value Chain and closer to the Consumer. So what information do we have that is valuable, and how do we use it?

Structuring Qualitative Data

In any Demand Planning process, there must be a quantitative way to evaluate and add information for new products, new customers, increased demand and potential upside, even though the sources of information are vague and qualitative. But without the benefit of POS data and Social Media, where does that information originate, and how does one quantify insight gathered from things like discussions? How do we decide if that information is correct? We can give a probabilistic value to quantitative data, but doing so with qualitative data is rather more difficult.

To add to the complexity, we included insight from past discussions on the probability of making additions to the product range. With differences of opinion about how certain one needed to be before changing a forecast to include speculative changes, this became a difficult process. Combined with a fear of wasting capacity, there was no clear process of how certain one needed to be that a sale would be realized before adding to the forecast.

This led to stakeholders using the forecast to hedge inventory for future sales, a methodology which may not be based on customer related information at all. This wasn’t what we wanted.

In turn, some forecasts generated error with high cumulative error and bias. This directly contributed to increased inventory and affected other metrics including delivery reliability, because the increased production of over-forecast products used up capacity that was needed for others.

The fear of wasting capacity, as well as pressure to forecast to capacity (whether logical from a sales perspective or not) caused further confusion. We clearly needed a solution that addressed the certainty of a sale vs. forecast. The solution also had to address the fears of exclusion as well as the effect of unreliable information on forecast accuracy. By using traditional forecast theory, our division was able to quantify information from discussions that previously presented challenges.

Gauging Certainty Of A Sale As A Function Of Time

Forecasts are less accurate as one moves forward in time, so it stands to reason there is less certainty in a forecast a year from now, than one that is used for production in the upcoming month or two. From this information, a guideline was developed to gauge the certainty of a sale to be included in a forecast as a function of time. (See Figure 1)

Figure 1

Forecasts in the near term are required to be 100% accurate, using confirmation from the customer. As we move further out in the timeline, a greater measure of speculation is allowed. The timeline was developed to be easy to use, easy to interpret by Marketing, and to produce accurate results.

Since we use one set of numbers, the inclusion in the operational forecast automatically meant alignment with the financial forecast. By establishing a process for quantifying insight gained from discussions and using Customer/Sales information, we no longer had to hedge inventory.

In the final analysis, the most experienced forecasters were following these guidelines all along. What was missing was definition and documentation. This clear definition of how uncertainty is handled in our forecast process not only benefited the less experienced forecaster, but changed perceptions for observers and users of the process, who thought the issue was inadequately addressed.

Conclusions

Even in the most robust of processes, small changes may have an effect on the overall results. At BASF, the way that certainty of a sale was viewed in the past was undefined and varied. Making a small change to the process, and defining the degree of certainty necessary to include in the forecast as a timeline, changed the way the forecast was developed and viewed. This exercise refined the way we will view process improvements in the future.

Alan Milliken - BASF

This past week the mighty LA Lakers, were swept by the Dallas Mavericks in four games.  The Lakers had won two straight NBA titles and were favorites of many to three-peat.  Immediately after the 4th loss, which turned out to be one of the five worst losses of  series deciding games in the history of the NBA, rumors began to circulate about conflicts within the organization.

Several experts said that fans should expect to see many changes within the Lakers Organization coming soon that would affect members  from top to bottom.  They commented that LA had not reached a consensus in regards to what is the best strategy or the best way to approach the execution of that strategy. .  There appeared to be disconnects between upper management and the coach as well as between the coach and the players.

The LA Lakers example exemplifies the fact that   it might be easy to bring a group of professionals together and say they are a team but achieving teamwork that leads to overall organizational success is much more difficult.  This  also makes it clear that spending money and throwing resources at a process does not guarantee success.  Quite often in sports when the under-dog wins you hear folks say, they may  not have the best resources but they had the better team.

The same principles apply to businesses and the S&OP process.  A strong S&OP Team can be the difference between winning and losing.  Conflict among team members and lack of consensus on the game plan can lead to disaster. Failure to gain commitment from all key stake holders can derail the process.  LA’s bench (reserves) were totally outplayed by the Dallas reserves.  We must always remember that operational excellence is driven by People-Process-Technology but only one has the ability to think and act accordingly.

Strong teams must be able to anticipate change and respond to issues in real-time whether they work in business or in sports.  Dallas’s strategy was to post up outside 3-point shooters, move the ball inside and then pass outside to take the open 3-point shot and LA was slow to recognize and even slower to respond. . Likewise, S&OP cannot be effective unless team members can quickly change plans and execute  the new plan.  If your competitor has decided to take some 3-point shots you had better be able to quickly mount the proper defense.  Those who cannot do so will find themselves trying to explain their ineptness to their fans in sports or in the case of S&OP to the Board of Directors.

If you are wondering what happened to the Lakers or more importantly why  your S&OP Team is not winning, you should attend the Best-of-the-Best S&OP Conference in Chicago.

There you will learn:

• The latest technology trends with software providers on the leading edge.
• How successful firms leverage people-process-technology to improve performance?
• How to design planning processes and configure software to enable best practices?
• How to build teamwork and consensus that results in winning.

Attend the IBF’s Best-of-the-Best S&OP Conference in Chicago this June.

Alan L. Milliken,  Business Process Education Manager
BASF Corporation

Hear Alan Speak at:

Alan Milliken

Folks, we must constantly remind ourselves that “process” takes precedence over “technology” and technology should be designed to enable best practices. During the e-revolution many of the process best practices were first identified by those designing demand planning software. For example, the ability to identify exceptions to pre-defined rules in real-time and instantly issue alerts to planners. This best practice was in place before technology. However, technology is advancing and being embraced so rapidly that it often dominates our processes by default. That is, our people are not keeping pace with the knowledge, education, skills and abilities via objective training required to leverage the software. In many cases this has led to sub-optimal processes for planning and forecasting, where systems are wrongfully blamed for failure.

As the Business Process Education Manager at BASF I often ask my students which of the three components of operational excellence (People-Process-Technology) controls overall success? Unfortunately, some answer our ERP system. Many of us are asking; do we really run our ERP system or is it running us? The latest advancements in ERP software provide high power heuristics that can automatically calculate order quantities, safety stocks, maximum inventory levels, etc., based in part on demand forecasts. Priorities and trade-offs can be managed by the system. However, the objectives and associated rules are determined by people. Therefore, how effectively we deploy new processes and systems relates directly to the knowledge and skill of our people.

We must act now to avoid the “Terminator Syndrome” (Systems and machines taking over). Our people can make the difference in whether technology or process receives priority.  But, to do so they must be educated and trained in planning processes and then how the latest software enables planning. For example, if a Demand Planner does not understand how the software creates the forecast or the recommended safety stock related to forecast error, they cannot contribute to improving process performance.

We all agree that more emphasis is needed on what is best overall for the firm. For example, we want to leverage our S&OP processes to maximize the value added to the bottom-line. Software firms have responded by providing advanced tools that can simulate and optimize demand & supply plans across the business. However, we must ask if our employee skills are keeping pace, particularly in process best practices and quantitative methods such as time series forecasting, causal modeling, statistics, and more. Furthermore, internal collaboration and teamwork is now recognized as most important to business success since technology has enabled a much more integrated approach to planning. What is your firm doing to prepare employees to contribute in this new environment of the system and machine?

The Institute of Business Forecasting & Planning, IBF is the leader in providing S&OP, Demand Planning and Forecasting education & training. The breadth of the IBF program includes understanding these processes from a strategic, tactical and operational perspective. For example, their conference presentations include how firms are using demand planning to improve business performance as well as tutorials on how to better perform specific tasks. Participants learn both technical and personal competencies needed to succeed in today’s fast-paced, ever-changing environment. For the Demand Planner, the Certified Professional Forecaster (CPF) program provides the knowledge and skill needed to contribute to process design and system configuration.

— Want to discuss the latest technology trends with software providers on the leading edge?
— Want to hear how successful firms leverage people-process-technology to improve performance?
— Want to learn how to design planning processes and configure software to enable best practices?
— Want to prepare yourself to better contribute to your firm’s success?

Attend the IBF’s Best Practices Conference in Orlando, October 24-26, 2010

Alan L. Milliken, CFPIM, CSCP, CPF
Business Process Education Manager
BASF Corporation

About Alan Milliken:

Alan Milliken is Business Process Education Manager for BASF Corporation. He has extensive experience as a practitioner, consultant and educator in supply chain & operations management. Alan spent over 20 years at major manufacturing sites serving in production, logistics, process control, operator training, and scheduling. He has spent the past 15 years as a business process consultant and educator. Alan served as a subject matter expert on the teams that created the Certified Supply Chain Professional (CSCP) program for APICS and the Certified Professional Forecaster (CPF) for the Institute of Business Forecasting & Planning, IBF. He holds an engineering degree from Auburn University and an MBA in Management from Clemson University.

See ALAN MILLIKEN Speak in Orlando at IBF’s:

$795 USD for 3-Days including Advanced S&OP Forum!

]]> https://demand-planning.com/2010/07/28/the-terminator-is-stopping-our-demand-planning-forecasting-education-are-we-prepared/feed/ 3 https://demand-planning.com/2009/09/08/the-magic-of-better-demand-planning-in-orlando-and-i-am-not-talking-about-mickey-mouse-or-donald-duck/ https://demand-planning.com/2009/09/08/the-magic-of-better-demand-planning-in-orlando-and-i-am-not-talking-about-mickey-mouse-or-donald-duck/#respond Tue, 08 Sep 2009 18:56:31 +0000 https://demand-planning.com/?p=312

Alan Milliken

Many firms have no strategy for dealing with unforecastables and some even pretend that all items are statistically forecastable.  Statistical tests need to be used to identify those articles which cannot be forecasted with conventional techniques, such as intermittent demand and demand with high variances.

For example, low volume and erratic demand may indicate that an item should not be offered in the current product configuration or at a particular location.  A key question that must be answered in such cases is whether or not the product is profitable and to what extent? If the product is profitable, the corrective action may be to move the customer to a more popular package size or assign the customer to a different warehouse with more demand for the product.

So, I am not talking about Mickey Mouse and Donald Duck.  I am talking about the upcoming IBF conference in October where you can learn the “magic” of better demand planning.

You will have many opportunities to learn how to improve both the quantitative and qualitative aspects of the Demand Planning process.  I will be presenting on how the world’s leading chemical company, BASF, develops strategies to improve forecasting and better manage unforecastables.  This includes methodologies for aligning product offerings with forecasting strategies and determining when conventional statistical tools are the right answer.  Mine and other presentations will provide you with the knowledge and understanding to optimize the interaction of people-process-technology within the demand planning function.

If you are looking for a little “magic” to improve your demand planning & forecasting, join us in Orlando.  You will hear about the use of volume & variance analysis to develop demand planning strategies and much more.

Of course, your comments here would be greatly appreciated.  How do you handle unforecastables?  It would be great to hear from you.

Alan L. Milliken
BASF Corporation

See ALAN L. MILLIKEN Speak at The IBF’S:

$695 (USD) for 3 Full Days!

October 12-14, 2009
Orlando Florida USA