## Is there a correlation between Inventories and Forecasting?

September 3rd, 2013 by Dr. Chockalingam

Some say yes………. and some say no!

There are many things questionable about Statistics and Modeling and of course, the famous or infamous, Normal Distribution.  Those who question the value of forecasting invariably point out that it is gravely inaccurate (and unfortunate) to assume that your demand is normally distributed.

Is Normal the tendency to be perfectly normal?  Or should we accept approximate normals and just only look for those that are perfectly Non-Normal?  Either your glass is half-empty or half-full!

So that we can come to meaningful conclusions, I am going to assume life is generally normal.

The classic formula for Safety Stock states thus:

Safety Stock = Service Level Constant * SQRT (Lead Time) * Root Mean Squared Error of (Demand vs. Actual),

assuming you are using a Demand Forecast to produce and stock inventories.  There is a lot of debate as to what to use as the Demand Forecast Error.  Debates range from dismissing the Fitted Forecast Error (in-sample forecast error) to the question of whether we can even establish an Expected Forecast Error.

In any case, using the standard deviation of actual historical demand will over-state the Safety Stock in a majority of cases.

In the current state, you may be carrying inventory to compensate for poor demand visibility besides what you need for demand over lead time.  There is a portion of inventory you are carrying for safety stock.

Any improvements in Forecast quality can be related to improvement in Safety stock levels, if we hold other things constant.

For simplicity sake, let us assume that your committed service level is 98% and your lead time is 2 months.  Then the above formula reduces to the following:

Safety Stock = 2.05  {Service} * Sqrt (2) {Lead time} * Expected Demand Forecast Error

Safety stock = 2.05 * 1.414 * EDFE (expected demand forecast error)

Safety stock = 2.90 * EDFE

In simplistic terms, the above means for every unit reduction in Error (Note Units not %), Safety stock will decrease by 2.9 units other things remaining constant.

If we can assume what our current level of forecast error is and what is our expected forecast error or target improvement is, we can establish a precise benefit in the form of safety stock reduction.  This assumes there is no forecast bias.  If there is a forecast bias (generally there is in organizations new to Demand Planning & S&OP), then there is another sizable reduction possible through the amount of inventory carried over lead time by reducing the bias.

Getting back to the no-bias assumption, what does it mean for a percent point improvement in WMAPE?  What is the percent improvement in safety stock for every point reduction in WMAPE?

This entirely depends on what your current level of forecast quality is.  If your current WMAPE is 50%, then every % point improvement in MAPE will result in 2% reduction in Safety Stock.  Please see the chart that shows the reduction in safety stock for every point reduction in MAPE at different Starting points.

If your current WMAPE is 60%, then every % point improvement will result in 2.5% improvement in safety Stock.

At higher levels of Forecast quality, it is difficult to make further climbs in forecast accuracy.  So any marginal improvement in Forecast quality with a higher starting point will result in more than proportionate benefits in inventory reduction.

There are other routes to go as well:

1.  Lead time Reduction – Reducing the lead time will result in Safety Stock as well but not as much compared to the reduction in Forecast Error.

2.  Variability in Lead time – If our supply is quite variable this has a much larger impact on Safety Stock.  This is not in the classic formula but the extended formula uses variability in lead time.

In practice I have seen this being mis-used and abused with very poor proxies for variability in lead time.  In an uncertain world of demand, it is important have better control over our own supply and schedule adherence.

We will cover the topic of safety stock in a special session in our workshop on September 18-20.  Please learn more details at

http://www.demandplanning.net/demandplanning_tutorialMA.htm.

## How to create models for weekly Forecasting in SAP APO – A Primer

May 5th, 2013 by Dr. Chockalingam

Generally demand planners forecast at the monthly level since forecasts are well behaved and seasonality is easy to identify in months.

But there are occasions when you may need a very good weekly split of the forecast. This can be achieved through a combination of approaches in APO DP

1. Proportional forecasting
2. Using historical weekly splits as a reference keyfigure or
3. Statistical modeling.

In our up-coming hands-on workshop on May 24, we will be adding a brief session on modeling at the weekly level so good intra-month splits can be achieved for the purposes of Production Planning and Detailed Scheduling (PPDS).  We will evaluate the usefulness of weekly models to achieve good weekly splits of the monthly forecasts.  What is the incremental value add of this process compared to using the APODPDANT proportioning keyfigure to derive the splits.

As we enter the last week of registration, only a few seats are left at rush pricing:

May 24th ‘Modeling & Metrics in SAP APO DP 1-day Workshop’ Boston, MA \$995
http://demandplanning.net/modeling_metrics_in_apo_2.htm.

This is an inter-active workshop facilitating you to work in your own SAP APO implementations under our supervision so you can test models on the fly.

May 22-23 ‘Demand Planning and Forecasting Workshop’ – Boston, MA \$1,295
http://demandplanning.net/demandplanning_tutorialMA.htm

Companies that have been in attendance include Bose, BASF, Novartis, Ghirardelli, Michael Foods, Newell Rubbermaid, Brown-Foreman, Spectrum Brands and many others.

SAP SCM APO users will hugely benefit by attending all three days of the workshop at the package price of \$1,995.  Group of three or more can get additional 10% discount with one payment method from the same company.

I look forward to discussing weekly forecasting and modeling with you in a few weeks!

## Planning Software – Part 1: How important is Integration?

April 21st, 2013 by Dr. Chockalingam

Continuing on this discussion on the important properties of a Planning Software application, I will discuss integration this week.

What is integration?  Why is it important?  How well should my planning tool integrate with the rest of the system landscape?

The word integration can be misused and misunderstood.  There are two different perspectives when it comes to integration.

• The first and most common one is from an IT perspective – Does the tool integrate and work well with my overall system landscape.   Integration is generally supposed to allows an easier implementation.  Many components come in pre-built to plug this into the enterprise system, so the re-work necessary to make data talk from one system to the other is minimal.
• The second one is from a business perspective – will the software allow me to integrate the business operations globally?  Does it allow the corporate management to see the aggregate or whole quickly every month to make decisions?  Does it also allow a global IBP process to drill down quickly into areas of specific issues and challenges and highlight them for potential future risks and opportunities.

For the business definition of integration, a distinction needs to be made between analytical work and reporting.  Most of the global decision making process probably needs a good Business Intelligence (BI) platform that will allow reported numbers to bubble up to global management and allow for what if queries and provide exception based reporting.  I do not believe any one is thinking about a super-integrated analytical platform where the CEO can question the Safety stock setting globally and make a change!!

*****************************

We will discuss the pros and cons of this criteria in our upcoming workshop with specific reference to SAP APO Demand Planning – Modeling and metrics – May Workshop

*****************************

Having ruled that out, then we focus on how the system can co-exist with other global ERP and Planning systems.  Do we want a different planning system in each country?  Or should every one be using SAP and speaking German?

I have seen this question being answered by the corporate head quarters.  The main business unit picks a planning platform and then inspire or influence the other local business units to move that platform over a period.

In summary, we should think about the following aspects while considering a planning tool from an integration perspective:

1.  Can the planning tool work with the ERP backbone to get data in and out easily with minimal custom interfaces?  Data can be directly sourced through the ERP tables in some applications without going to the data mart or data ware house.  However in many practical instances, there may be customizations necessary to extract data even with in the same ERP/planning software family.

2.  Does it fit with the overall system landscape in terms of the touch and feel?  This may be a criterion for the simple reason of user adoption.  The interface and familiarity of the user interface will make it more easily saleable and trainable to the user.

3. How does it relate to other planning applications used?  When the demand forecast is passed to the Supply planning system, do both systems interpret the forecast to mean the same thing – units, detail, horizon, expected forecast error etc.

4. If it is a best-of-breed tool that is NOT from the ERP family, how easy or difficult is it to build the integration capabilities to the ERP system?

In summary, integration may be one of the important criteria.  But is this a critical criterion to the exclusion of others in choosing a planning software?

## Planning Software – What should it be?

March 10th, 2013 by Dr. Chockalingam

Having worked with many software applications for planning – demand, supply, finance, and S&OP – I have experienced a sense of elation (seeing something work very nicely) and shock (seeing something that work so stupidly on something so obvious).  In a series of blog entries starting this week, I am planning to examine the important characteristics of planning software.

Broadly, we need to think about planning as an activity that involves both analytics and judgement.  Although partly this is a decision science oriented activity, ultimately a planning activity should involve the planner to make a decision, a judgement call.

In Operations Research, a planner may put together a detailed optimization problem – Decision Variable, Cost variables and various  Constraints and finally will pick the alternative that has satisfied the constraints and produced the maximum value for the decision variable or minimum value as the case may be.  Judgement may still come in when they want to alter different variables and constraints.

Although my focus is going to be predominantly on business forecasting – Demand Planning, Financial Planning and S&OP, I think we can use the same characteristics to evaluate many other areas as well.

There are many things that people use to evaluate software packages to see if they fit their business need.  In reality, the businesses are under-prepared in the evaluation activity and they let the software vendor’s marketing lead them forward.

What are the “many things” that people think about when evaluating software packages?

1.  Cost – Can we afford it?  Is it the cheapest among the contenders?

2.  user friendly – Is it more like Google (where the user types the question and gets the answer) or is it on the other extreme – where it takes an year to read the user manual and get trained?

3.  Clarity – Do I know the reason and the math behind the analytics?  Or is it a black box?

4.  Speed and Performance – Do I get responses quick or I wake up dreaming about the sand clock?

5.  content – Can it solve our business problem or is it just a typing tool?

6.  Reporting – Can I produce good reports from the software?

The usability is a key component of the evaluation matrix.  Both users and management would like the planning tool to be easily understood and to allow the planners to do the job right.  Ideally the software should work for you instead of your laboring over it to produce what you need to produce.

What are some of the key characteristics of a planning tool that makes it usable and valuable?

Let me get to the specifics of a forecasting and planning application now.

Feature 1:  Integration

How well should my planning tool integrate with the rest of the system landscape?

Integration should be part of the evaluation matrix although not necessarily the main requirement.  From an IT perspective, integration allows an easier implementation.  Many components come in pre-built to plug this into the enterprise system, so the re-work necessary to make data talk from one system to the other is minimal.

But integration is also important from another perspective – it should provide a seamless user interface.  The look and feel of the planning software is not very different from the rest of the system landscape.  Many gadgets that work with your enterprise system should continue to work with the planning software.

Ctrl-C and Ctrl-V refer to Copy and Paste in Windows OS.  They work pretty much every where with all applications installed and are compatible with Windows.

In SAP, the interface is seamless – APO and ECC work together.  Typing different T-codes will get you to different areas but the user may not even know that they are using different modules.

How valuable is the feature of integration?  Is it worth paying the price of usability and content?

We will talk about the second characteristic next week.

Sincerely,

Mark Chockalingam

Chief Demand Planner

Demand Planning LLC

## Demand Planning Training Workshops – What is out there? What is different about us?

January 11th, 2013 by Dr. Chockalingam

There are various queries on Linked-In about the best courses in Demand Planning and Sales Forecasting.  It is time to set the record straight on good Demand Planning Training workshops and why Demand Planning LLC is the best provider for Demand Planning and S&OP training workshops.

I have seen two types of offerings in the market place:

1. Extremely academic and Formula oriented workshops – Good for the academic researcher but of little practical value to the Demand Planner.  These courses focus heavily on the math and talk about serial negative auto-correlations, heteroscedasticity, Generalized autoregressive conditional heteroskedasticity, ARCH, GARCH etc.  Not that there is anything wrong with it, it may definitely be useful while getting a Ph.D. in Econometrics.

1. Concept oriented courses that cover best and worst practices – and more often worst practices – but these workshops limit themselves to concepts.  They do not actually help instructing people on how to do this better in their every day jobs.  I have heard more people say that some of these workshops all talk about what is not working all the time.  I have also seen really light-weight courses that use a lot of lingo and lot of citations of real stuff happening in the business world – but without much numerical examples or calculations or problem-solving.  Concepts are good but without no test of the theory in practice, they are not good enough.

The workshops offered by Demand Planning Net are unique in the sense we actually offer practical, hands-on workshops.  We assign pre-work to every attendee before the come to the workshop.  Every one needs to bring a laptop to work through individual and group exercises to create actual demand plans and solve demand analytic challenges.

They submit solved case studies before they get a certificate of completion.  This certificate of completion is essential as a pre-requisite before appearing for the certification exams in Demand Planning and Sales Forecasting.

Many professionals who have attended our workshops, have reported back that they were able to use the concepts in their every day activity.  Many of our attendees are our loyal fans and have sent their colleagues and peers and successors to the workshops.

Here is a comment from a planner who was new to the role:

“I am new to this field and new to this type of analytics. I found this workshop to be one of the best things my company could have sent me to. THANK YOU! ”

“Out of all demand planning seminars, classes, speaking events, and conferences I found this workshop to be one of the best things my company could have sent me to. I would recommend this to anyone, even veterans in the field.”   – Demand Analyst at RedGold

Here are couple of video testimonials-

You can review all the testimonials at

We understand you have a choice in the market place on what courses to attend and how to judiciously spend your company’s training budget.  A little bit of research would go a long away along with questioning your objectives. What is important – theory, concepts, practice or a combination of things that actually help you be a better planner.

The entire mission of our firm Demand Planning Net is to make planning better and better the planners!

You can attend our Dallas Workshop in Feb 2013 or the Boston Workshop in May 2013.

http://demandplanning.net/workshops.htm

Please call my office with questions!

## Coming to Oil & Gas Country – Demand Planning and Forecasting – Texas Feb 2013

October 16th, 2012 by Dr. Chockalingam
We will be hosting our first ever public workshop in Texas due to popular request from our Oil & Gas colleagues and planners at Oil Field Services companies.

Demand Planning Net will be bringing our popular two day tutorial workshop on Demand Planning and Sales Forecasting to Dallas, Texas in February 2013.  The two day workshop is scheduled for Feb 27-28 in Dallas, TX to be followed by our unique one-day seminar on Modeling and Metrics in SAP APO Demand Planning.

We just concluded our Sep 2012 workshop with a record turnout of brand name companies including Eaton, Michael Foods, Schlumberger, Thermofisher, Newell Rubbermaid, Lexmark, Abbott and others. Read our past attendees testimonials at – http://demandplanning.net/testimonials.htm

For course content and registrations please visit http://demandplanning.net/demandplanning_tutorialCA.htm

Why is this an important career enhancing workshop?
Attendees who complete the two day course and all exercises will be awarded a certificate of completion. This workshop will be the pre-cursor and required workshop to our certification program to be launched in 2012.

Get skills you can use at work
We will explain and demonstrate best practices in statistical model selection, illustrate how to improve model quality, and teach you how to leverage the forecast measurement process.

Learn from industry experience
We will bring practical examples from our consulting experience with clients in Consumer goods, Food and Beverage, Chemicals, Pharmaceuticals, Heavy Manufacturing, Aerospace, Medical Devices, Oil and Gas etc.

Network with peers
You will have the opportunity to meet, interact, and learn from other demand planning professionals with team challenges and networking exercises. Attending this workshop will introduce you to our vast network of supply chain professionals and career opportunities in North America.

Upon completion, you will be awarded a certificate of completion from Demand Planning LLC, attesting to your newly-acquired skills in Demand Planning and Forecasting.

Does the two day workshop contains topics related to forecasting of rarely used spare parts or not?
Yes we talk about models to forecast and stock for rarely used spare parts otherwise known as intermittent demand. While the focus is on forecasting and modeling, our approach is to marry up the business problem with the technology.

Look forward to seeing you in Dallas for our February 2013 workshop!! Early bird pricing is valid until December 1, 2012.

For course content and registrations please visit http://demandplanning.net/demandplanning_tutorialCA.htm

Contact me if you have any questions- 781-995-0685 or info@demandplanning.net

Happy Forecasting!

## Demand Sensing, Demand Shaping and Demand Management – Are demand planners doing these activities today?

September 26th, 2012 by Dr. Chockalingam
Demand Planning Net focuses on the disciplines of Demand Forecasting, Demand Planning and Demand Management besides the related areas of S&OP and Inventory Optimization.
Occasionally there is a confusion when people talk about Demand Management. Some companies and consultants call classic Demand Planning as Demand Management. Even some consulting companies call the planning function as Demand Management.The demand side of the business should be engaged in Demand creation, Demand sensing and demand shaping and finally the Demand Management. So Selling and Marketing functions are in essence Demand creators and Demand Managers.

Demand Planners should play a facilitative and analytical role in the core selling function of Demand Management. They have the ability to predictive demand. They should also have the ability to understand causal effects of demand drivers.

So I see the Demand Planning function as an important partner in the Demand Management role.

Do the modern day demand planners help the company in

* Demand Sensing
* Demand Shaping
* Demand Management

I have seen Demand Planners reporting to the Supply chain function only do the last role namely as Demand Managers when there is supply shortage. They act as the bearers of the bad news to both Sales and Customers and work on rationing product.

On the other hand, Demand Planners reporting into Sales, Marketing or Finance generally play a more active role in some of the Sales Analytics function. They are also called upon to measure promotional effectiveness and pro-active promotional planning.

Have you Demand-Shaped? Has your VP of Sales asked you a what-if question in the last 30 days?

By the way,

We have additional details on our consulting services in the Solutions Area.  Look at the detailed Solutions consulting page at http://demandplanning.net/solutions.htm.  There are four broad areas of solutions consulting:

• Usability Consulting for tools such as JDA Manugistics, i2, Oracle Demantra and SAP SCM APO Demand Planning
• Re-design and Re-implementation services for SAP APO DP

## Keep Cribbing or Find Solutions

September 2nd, 2012 by Dr. Chockalingam

There was an interesting technical issue at a client using SAP SCM with substantial investment in demand planning using SCM tools.   After some googling, I landed on this blog site written by a consultant who obviously makes a living on SAP.  But strangely enough, this consultant has created a complete web of blog entries filled with complaints about SAP and APO particularly demand planning.  The sole purpose of his writing seems to be scathing attacks on SAP.  Unfortunately there was not any content that could solve problems or answer questions.  Surely I did not find an answer to my question on his blog.

He does have pretty good complaints – very well articulated.  These complaints are not new about SAP APO.  Most users know these complaints and they live through this every day to their frustration.  I have articulated them before – some of the problems are tool specific that SAP has not bothered to correct.  It is a shame the company will not listen to the users or read the thousands of OSS notes written every day.  The problems that SAP has to be blamed for include issues such as forecast error calculations, incorrect alert logic, manual effort required in life cycle planning among others.

However there are many problems that are created by the Integrators/Consultants including our “esteemed” blog writer due to their lack of knowledge of both demand forecasting as well as an understanding of the tool.  If they don’t understand the tool, they should not be implementing it or training the users.  The baggage they leave behind creates a mess that makes the software worthless and unusable.  These problems include:

1.  Implement APO to be used as a typing tool and tell the users Statistics are terrible so not worth using it.

2.  Disable statistical modeling under the pretext of security.  In reality, the consultants are worried about fielding questions from the users on statistics that they do not understand themselves.

3.  Enabling options and parameters in the tool with out any idea of what they do to the resulting forecasts.

4.  Deciding on important things such as forecast aggregations, forecasting levels and exception management without any process discovery or user input.

5.  Finally not project planning the budget to include training to the users particularly on the Stats and Demand Planning.

Perhaps companies should divide the implementation plan into two parts – process design that includes the provision for model tuning and training and the second part that involves system design and making the system to work.  The system design should follow the process design.  The model tuning should be done by the same consultant that is responsible for process design.  And definitely that person should be an Expert not only in the tool but also in best practice demand planning along with expert skills in training the planners.

Given how many implementations are currently plagued with tool problems aggravated by Consultant inefficiency and incompetence, perhaps more companies should think about re-implementations of APO DP.  Just throw away the old concepts and practices and start thinking about how to fix the problems and make the tool more usable.

Finally SAP also needs to wake up and start fixing the problems in their most popular SCM module namely APO Demand Planning.  It needs to fix the error calculations and the alert logic.  More on that in a separate blog entry.

To set the record straight with this “esteemed” writer so he does not pollute the waters and mislead many planners, I would say the following:

The blog writer concludes that SAP APO calculates the MAPE incorrectly.   I agree with him to a certain extent.   SAP purports to compute MAPE using the academic definition of averaging percentages but it does not do this either.  It goes into a hole when the actual demand is zero and makes the MAPE metric unusable.  However, the other metrics namely MAD and RMSE are correct.

I strongly disagree with this Consultant/writer when he concludes that best fit models are erroneous because the error calculations are defective.  If you poke around the underlying mechanics which is well documented in the APO online manual, you will know that the optimization is done using the Mean Absolute Deviation, which is superior to the MAPE; Mape is a percentage and has some awkward properties.

The Automodel selection 1 uses the MAD and picks only smoothing models while the Automodel selection 2 also claims to include linear regression models.   But in practice Automodel selection 2 produces inferior results and expects the user to baby sit the modeling by feeding manual parameters!!  In general Automodel is not for the faint of heart as many settings have to be correctly configured.   Given the fact most configurations are done by junior consultants from big 5 consulting firms, you can guarantee that this is an unrealistic assumption.

Even with other forecasting packages we do not recommend best fit models or expert selection as the final model.  They are good starting points, but the planner has to do more in getting to the right model and the demand forecast.   They don’t have to be expert statisticians but they need to understand their business and have a preliminary understanding of what various models do.

Yes the Statistical models are straight forward in APO, in fact, they are basic.  There is no complexity in them.  They are not claiming to do Box-Jenkins or Transfer functions or ARMAX or any other models with esoteric names.  However, I have found people use the MLR models very cleverly combined with forecast attributes.  So it is all in implementation, model tuning and finally imparting that much elusive knowledge to the planners and find a way to sustain that knowledge.

My two cents = Do what you can and understand what you cannot.  Some intellectual honesty will also go a long way!

Happy Labor Day weekend!

## Forecast Error Benchmarking across various industry – survey results

August 22nd, 2012 by Rohan Asardohkar

· Which metrics should I use to measure my forecast performance – WMAPE, MAPE, Bias or something else?

· Is there any benchmark available for forecast error, particularly within my industry?

With this in mind, this past Spring we started conducting the survey across supply chain and demand planning professionals from various industries. This survey was meant to compile information about their pain points, forecast error metrics they use, industry they work for, and who owns the demand planning function. We are publishing the first installment of the results from this survey in this newsletter.

As expected one of the metrics used by 52% of the respondents is WMAPE or volume weighted MAPE, calculated as Sum of Absolute errors divided by sum of actual demand.

We have good informative data for Chemical, Consumer Goods (CPG) industries with a good sample size and participation from a broad range of companies.  CPG in our study included Food and Beverages as well.  For CPG industries average of forecast error is 39%.

One of the common pain point for CPG folks is number of variables such as price fluctuations, promotion timings, and new items. For Chemicals industry the average of forecast error is 36%.  Promotions are not formally planned or executed in Chemicals as in CPG, however there may be price incentives etc.