Peter introduced Roxanne to the team members, although most had already met her. Since there were a few changes following the meeting in the Marriott Marquis, Peter asked the team leaders to introduce themselves and explain their role in the company and on the team.

Peter explained that his Planning and Control Milestone Team was responsible for the Supply Model, which combines the Planning Spine with the geography of plants and warehouses. “As you know, I’ll also be attending the Integrated Business Management Team meetings to understand how the detailed planning processes fit with the aggregate. On the list [Figure 15.1], you’ll see that Janice’s name also appears next to mine. She’s going to be a key player in all aspects of supply planning. Janice, I’m going to rely on your expertise a great deal. This means Janice will average about 50 percent of her time on this work; but that will vary from as much as 100 percent for a few days down to just 20 percent at other times. Janice will speak for me in my absence. I’ll keep in close touch with you, Janice, so we’ll always be aligned. This will also ensure that the planning model and our designs stay aligned with each other, while giving me time to deal with both intra- and inter-milestone issues.

“We’ve developed an Output-Gantt chart that shows all the outputs as deliverables from the tasks required to achieve team objectives. The Output-Gantt is quite simple to follow. All input tasks are excluded for simplicity. These are driven by your personal Input-Gantt charts and your commitment. As a reminder, ‘Silence-is-Approval’.

“Over the next few weeks, we have some design team workshops scheduled. We start with a two-day workshop facilitated by Roxanne on the Supply Model; that’s already scheduled. Core members, as indicated with the ^b against your name on the team list (and in Figure 15.1), must attend as well as the rest of you if at all possible. Most design activities can run in parallel since we all know what our deliverables are. The Output-Gantt chart shows them and their due dates. Have I said before, ‘Don’t miss your due dates’?” Everyone laughed at Peter’s unnecessary repetition of his favorite message.

“I’m suggesting we have a abbreviated name that we use for our Capable Planning and Control Milestone work. How about the acronym, ‘CPC,’ standing for Cosmetics Planning and Control, but easier to say? What do you think?”

“I like the idea of an acronym,” Janice said as the rest nodded approval. “I think it might be even better,” she continued, “to call this the Customer Service Team (CST) conference room to keep focus on our real objective.” There was unanimous approval, and the new name was adopted.

“Great; I’ll have a sign made for the door. Now let’s get back to the schedule. I’ve also scheduled the Inventory Accuracy workshop with Dan and the Data Integrity workshop with Tom. The Design Leaders, Francis and Sally, have made sure that all their design team members are available. Francis is including some shop floor team leaders and lead operators from the warehouse. Gordon Fast from Finance will be there to make sure we incorporate the needs of the Finance group. Bob and Chris, you’ll be attending the Data Integrity Workshop with Sally so that you can see where we’re headed from the standpoint of documentation and IT support. But don’t just observe. We need your best thinking about this subject, too. I’ll confirm the schedule for the other workshops next week.”

Peter then handed the meeting over to Roxanne. “Thanks, Peter. I’d like to explain what we at Effective Management mean by the term ‘workshop.’ Workshop means that you do the work, and I facilitate. I’ll bring along some slides to help you get organized and to remind you of key concepts and principles that must be built in to your designs. Please bring along your course notebooks for reference. We’ll spend 20 percent or less of the time on reviewing the educational material, then get down to the design work.” She explained that the main objective for the next six to eight weeks was building the design for the Planning Spine [Figure 15.2] and enablers such as inventory accuracy and data integrity processes. All process design team activities will be directed toward completing this model.

After a brief review of the Planning Spine with the promise of more coaching to come, Roxanne asked Peter to take over. He turned to the first objective of the team, building the Planning Spine.

“Let me recap. We’ll be leading the process design teams’ work and integrating their proposed designs into a supply model for our business. The process design teams’ outputs will be put together and tested in our ERP system. As I understand it, when we get it right, we’ll have the optimized ‘input-process-output’ elements linking customer needs right through our supply chain and into the suppliers’ plans. We’ll test all designs before they go into the model, and we’ll be careful not to say ‘no’ to an idea before we think through it, try it, and provide feedback to the team. I want all the teams to know that we’re listening. One exception. At no time will we violate the key principles and concepts we’ve learned in the detail courses and workshops. Part of my job is to keep us all true to the Class A concepts and principles. Using our own ERP test database will allow us to make sure we’re integrating all the designs and allow us to test creative possibilities. We’ve actually set up three database sets to work with. Later I’ll give you new secure sign-in details.”

Gordon interrupted, “Why three database sets?”

“Thanks, Gordon; good question. We need one to build our Planning Spine covering the internal supply chain and interfacing with the outside world. This is the real CST database. Only tried and tested stuff goes there. The second database, CST-Test, is where we’ll develop and test alternatives. With David’s approval we’re going to learn by doing, rather than risk ‘analysis-paralysis.’ This is the dataset we’ll use the most. The third dataset, CST-Train, will be used later to support all the training we’ll need to do as we complete the design. I expect we’ll also develop some CBT modules to help with the training.”

David raised his hand, “Better explain CBT, Pete.”

“Okay. CBT means computer-based training. It allows people to use their computer work stations independently to learn at their own pace how we’ll be using the ERP system to run the business in the future. The CBT modules will provide a learning library for new people, for ongoing refresher training, and to help roll out process changes.” Peter asked Roxanne if she’d like to add anything.

“Thanks, Peter. I do have a couple important points. We need to standardize our terminology during implementation so we know we’re talking about the same things. First, we need a working definition of the words ‘Planning,’ ‘Scheduling,’ and ‘Execution.’ This is what I suggest:

“Planning is about the future. It defines what we require to happen to meet the Master Supply Plan at all levels from end-item to supplier requirements. It commits the supply chain to provide the resources (people, equipment, facilities, and, of course, money) to support that plan. However, it does not result directly in things being produced. The calculations are done automatically by the Enterprise Resource Planning (ERP) system utilizing user-defined rules based on inventory records and master data, such as item, routing, work location, and bill of materials masters. The master supply planner has accountability for the overall validity of these integrated plans.

“Scheduling is the process of sequencing specific operations at each work location, within delivery and capacity constraints. All schedules must be doable and are reviewed for validity at least daily. Scheduling starts when a Material Planner releases an order, mostly prompted by the ERP system through action messages. This work order commits the supply chain to produce a specific product and quantity at a specific time. From this point on the order is under control of the material and capacity planners, not the system. The output is a detailed Production Schedule by work location. The Material Planner owns and controls the start and due dates of all released orders. The Capacity Planner has the ability to override the operation due dates but cannot override the order due date.

“Execution means working to the detailed Production Schedule or Supplier Schedule to supply the ordered product. The shop floor (company’s or supplier’s) follows sequentially the work as shown on the detailed Schedule, updated at least daily, to transform incoming materials and components into the scheduled supply item, thereby meeting the overall plan.

“As you can see, there is a time-based flow from Planning → Scheduling →> Execution.”

Peter added, “I see what you mean. If you’d been talking about scheduling, I might well have been hearing planning. I like the clarity of your definitions. For the moment, team, I suggest we adopt these.” Everyone agreed. “Any more useful definitions, Roxanne?”

“Sure, Peter, you’ll see some of these on the Planning Spine model—the planning roles. I’m going to describe planning roles for Capable level. But the closer you move to a ‘flow’ supply environment, typically in Phase 2, the more you’ll find that these roles become combined and managed by one individual or a small team focused on the end-to-end flow from supplier to customer. But all our Capable level reference material and coaching relates to these roles:

Peter interrupted again. “Looking at the Planning Spine model [Figure 15.2], two questions come to mind, Roxanne. First, why are the Demand Planning Manager and Supply Planning Manager shown next to each other on the model? And second, where’s the Master Production Scheduler role in all this?”

“Excellent question, Peter; I’m glad you asked, but let me answer your questions in reverse order. The Master Supply Planner might well have been called the Master Production Scheduler in the past. But the title is no longer appropriate when you consider how the responsibilities develop as a company moves from Capable Planning and Control to Advanced and then on to Excellent. The title ‘Master Production Scheduler’ is historically supply-side focused and brings to mind the old days when manufacturing’s motto was ‘you can sell what we make.’ Hardly customer focused!”

Peter replied, “But that’s exactly how we seem to plan today given our long production cycles. It just might help explain part of our customer service problem! I agree that today’s planning model isn’t very customer focused.”

“And there’s more, Peter. Remember our definition of scheduling? Scheduling covers the short- to medium-term planning horizon. But today, companies must create supply plans covering the next 24 months and sometimes even longer, so the old term is limiting. I still like and use the term ‘Master,’ as in Master Supply Plan. This takes us back to your first question. The Master Supply Plan is developed by the Master Supply Planner, as the item-level expression of the Supply Planning Manager’s aggregate Supply Plan. It drives all other plans, but it is co-owned by the Demand Manager. That’s a new concept for most companies beginning this journey. Many of my clients—usually the best at customer service—have the Supply Planning Manager and Demand Planning Manager co-located so that they can work closely together in developing the optimal Master Supply Plan for the business. And there needs to be a Master Supply Planner at each facility with the responsibility to support their facility’s supply plan as assigned by the Supply Planning Manager. An alternative to co-location used by some companies, almost as effective, may mean using technology so they can collaborate as though they were actually sitting together. That’s why there’s no space between the roles in the model.”

There were disgruntled looks on the faces of the plant master production schedulers, but Peter decided to reinforce Roxanne’s message to head off any negative comments.

“That design will change a few things and will take time to get used to. But I can see some big advantages to that planning model. After all, manufacturing has no purpose other than to serve marketing does it?” As he studied the faces of his team members and recalled the positions taken by Tony, he suspected a paradigm shift of that magnitude would take persistence and coaching to achieve. He continued. “As we begin to fully understand and appreciate the benefits of this change, I know we can get people aligned and supportive. It’s absolutely consistent with the focus and integration that David’s been talking about all along.” With that encouragement from Peter, Roxanne continued.

“While the Master Supply Planner establishes the Master Supply Plan and responds to the requests for short-term demand changes from the Demand Manager, the person in this role must also respond to customer orders. These changing demands often require decision-making trade-offs. As robust as your new planning processes will become, you’ll always encounter the need for trade-offs between customer service, inventory, and costs; it’s just a fact of life, especially in Phase 1. But you’ll soon see that as your processes improve, the impact of these trade-offs will diminish noticeably. Well talk more about this later in your implementation.”

“Let’s move on to the next planning level. Material Planners are accountable for managing the item ordering rules in the master data files for inventory and safety stock levels, and for responding to ERP system action and exception messages to synchronize the internal supply chain. This is the key day-to-day objective in maintaining material plan validity—the ‘matched set of parts’ term that Oliver Wight coined some years ago. Before you get the wrong impression, I had better qualify what I just said. The Master Supply Plan automatically feeds MRP, the Materials Requirements Planning calculator within your ERP system. In a Capable Planning and Control environment the MRP software calculates a complete, valid material plan from end item to supplier requirements, all synchronized and all conforming to your optimized rules to support your Master Supply Plan [Figure 15.3].”

“We run MRP today,” said Sandy Bar-Nestor, “but you make it sound a whole lot better than it is.”

“There’s a reason for that, Sandy. You’ve never really built a working Planning Spine, the sum of all the integrated business rules governing how you run your internal supply chain. You’ve never exploited the capabilities of your ERP system because your current business processes aren’t designed for integration. They’re designed for functional optimization, or even worse, functional reaction.” Sandy didn’t look convinced, but allowed Roxanne to continue.

“Supplier Planners, often called Supplier Schedulers, are the day-to-day points of contact for your suppliers. They maintain material plans in MRP by supplier and by commodity. The output is scheduled by supplier by item/part number. They maintain valid supplier schedules that summarize requirements by supplier across the entire planning horizon. They are often matrix-managed, solid line to planning excellence; dotted line to purchasing. In some companies with a low volume of call-offs from suppliers, these two roles may be filled by the same person. The Supplier Planner works within the commercial contract, negotiated by the Purchasing or Procurement organization, between the company and the supplier. If you consider the contract as the long-range plan, then you can consider this role as scheduling. It includes scheduling inbound deliveries, ensuring deliveries arrive on schedule, and monitoring supplier delivery performance on behalf of the buyer. The primary tool for communication between the Supplier Planner and the supplier is the Supplier Schedule, a direct extract from the material plan but organized by supplier, rather than by product stream. All Supplier Schedules directly extracted from the material plans are preauthorized by the agreed-to Master Supply Plan and do not require secondary authorization, approval signatures, or intervention by the buyer.” Pauline Powell, a purchasing buyer, interrupted at this point.

“Hold on right there, Roxanne—it’s company policy that all purchase orders valued over $5,000 must be referred to the buyer on a Purchase Requisition, so the buyer can authorize a purchase order using our skill and judgment before that money can be spent. I often change requisitions to better match the latest price breaks. I can’t accept this Supplier Schedule procedure. You’ll certainly support me on this, won’t you, Peter?”

“One thing I’m learning rapidly with this initiative, Pauline, is that the old saying is true, ‘If you always do what you’ve always done, you’ll always get what you’ve always gotten.’ And we sure don’t want to perpetuate what we’ve always gotten! While I suspect Roxanne is right, I would like a little more clarification and rationale, Roxanne.”

“Thanks Peter. Pauline, you’ve raised an appropriate and fairly common concern. The way Cosmetics Products operates today, I’m not going to argue that your current procedure is right or wrong. What I know for certain is that perpetuating your current procedures in your new design would be absolutely wrong. We’ll provide a day’s education on the details, and go through any concerns at that time. But let me ask you a question, Pauline. How much of your day is spent processing purchase requisitions and purchase orders, and in talking with suppliers about urgent or late deliveries?”

“I’d say about half of my time.”

“Good. How quickly are you able to process requisitions?”

“Well, obviously I have other responsibilities, but I make sure I clear my inbox by the time I leave work every Friday.”

“So, Pauline, it is possible that a requisition arriving at your desk late Friday could wait until the following Friday to be approved and a purchase order created and sent to the supplier?”

“Sure. That can happen, but it doesn’t happen often. I usually take care of them within a couple of days.”

“Do you consider carefully screening purchase requisitions and purchase orders as a valuable use of your purchasing skills and abilities?”

“Well, it is an important requirement within our procedures, but in terms of skills, it doesn’t require much. I could contribute more to the business by spending more time developing our suppliers’ capability to support Cosmetics Products. But that’s not possible. I have to get the purchase orders moving to support production.”

Roxanne now broadened her response as she turned to the group. “I need to remind all of you that I am not criticizing the way you work today. I know you are all working hard and doing your best with the procedures that exist today; we all have to accept that. But, as with many of your processes, there are much more effective ways to do things. At some point it won’t be Pauline I’ll be having this conversation with, it might be you. So don’t think Pauline is alone here.

“Now here’s the brief answer to your question, Pauline. First, the annual Business Plan is authorized by the senior executive. Then each month, the aggregate Supply Plan supporting that Business Plan is updated and authorized during the Management Business Review. In preparation for the Management Business Review, the Master Supply Plan is reviewed and approved each month during the Supply Review, and adjusted as needed weekly as part of the weekly Master Supply Planning meeting. Additionally, the Master Supply Plan is maintained daily so that it is always valid, with no ‘past-dues,’ and it continues to aggregate to match the authorized Supply Plan. To support the detailed material plans, you will develop authorized Cosmetics Products’ business rules. These will define precisely how you want material plans to be calculated and managed. Finally, the Material Planners are responsible for following the order replenishment policies and managing inventory to the authorized levels. Providing that the replenishment order is a direct flow through MRP to the Supplier Schedule, the material release has already been authorized on several levels. There is no need to involve the buyers; there is no need for a purchase requisition; and there is no need for approval of such purchase orders. They become nonvalue-added and redundant pieces of paper. So, Pauline, this frees up your time to develop the suppliers; an activity that is much more rewarding than chasing paperwork!” Pauline nodded, though still concerned about how to get agreement to change the current purchasing policy. “If we’ve covered that question sufficiently for now, I’d like to continue.

“Let’s move on to the role of the Capacity Planner. The Capacity Planners are responsible for ensuring capacity constraints are identified and managed to support the production schedule while optimizing efficiency. The cardinal rule is to meet the order due date. The planner analyzes required and available capacity with the owner of the resource. Analysis is accomplished using the ERP system’s detailed capacity planning module to determine the order’s operation start and completion dates based on the routing. Capacity Planners may override the operation schedule and select alternative routings to meet the order dates or improve efficiency. The output is a detailed Production Schedule by work location.”

Peter declared a 30-minute break for lunch, to everyone’s relief. Working through the details of terminology to create a common language, while essential, was proving to be tedious.

After the break, Peter addressed the group. “Now that we’ve been through the first education sessions, I know most of you have questions or concerns. I’ve set aside the next three hours to address those questions and concerns. Here’s your chance to get all your issues on the table.” The lively session that followed was a combination of education and team building. Between mini-lectures, Roxanne observed behaviors. She began to develop a sense of who tried to dominate, who was behaving negatively, who was not engaged, and who was silent, yet engaged. She also observed and was impressed with Peter’s leadership abilities, critical to the success of the Customer Service Initiative. At appropriate times, he was assertive; at other times he drew in the quieter members; he sought alternative perspectives to negative comments; and generally kept team behaviors appropriate. He was, as Greg assured her, a good choice to lead the teams.

The meeting with the Planning and Control Milestone team enabled the design teams to begin their work. Roxanne coached them to seek help quickly when they ran into a barrier. “Try to resolve your issues using the concepts and principles you’ve been learning. But don’t waste too much time in debating alternatives. If you have questions about the best approach, talk with Peter. He’ll call me to help if necessary.”

The Process Design team leaders arranged their own design team work plans, all beginning the week after the milestone meeting. Peter set a deadline of six weeks for each team to complete the design for their portion of the Planning Spine model, recognizing that implementation would begin when the design was approved by the steering committee. “Here’s a simple requirement; don’t be late! I’ll be around to remove any roadblocks, assist in answering questions, and arrange for periodic visits so that Roxanne can address your questions and review your progress. In return, I expect you to tell me immediately of potential problems. That will allow us to shift resources around if necessary to stay on schedule. You need to know that I don’t respond well to schedule surprises. Don’t count the hours; just get the job done.

“By the way, thanks to each of you team leaders for getting me your output-Gantt charts. With the few adjustments we’ve discussed over the past week, they look good and fit the overall time line.

“Looking at the schedule, our next meeting is our regular weekly meeting on Friday at 8:00 A.M. This week we’ll select four or five representative finished products and create the master data for them. Then in our meeting on the following Tuesday, we’ll review what we’ve learned and split up into teams working in parallel to finalize bills of material (BOM) structures and item master data. We’ll then enter the bills and item masters in the ERP system test database and run system reports to check how everything fits together. I can tell you I’m glad we don’t have to implement a new ERP system at the same time we’re doing this work. Roxanne tells me that lots of companies have to do that at a cost of many months added to their implementation schedule.

“We’re also going to kickoff all the other teams next week, except for the Supplier team. We’ll need to have our own act together, meaning have valid and stable supplier schedules, before we approach our suppliers.” Peter fielded a few more questions and adjourned the meeting.

On Friday morning, the Capable Planning and Control Milestone team members received their secure login and security codes for the three newly created database sets. With access to ten new workstations, they selected five major SKUs that were representative of all the key types of products. They split into pairs and began collecting BOM and item master data for each SKU and component material. Each pair reviewed every ERP item data field for appropriateness, paying particular attention to validity of order quantity, order multiples, lead times, and safety stocks. It was immediately obvious to them that the master data had never before been reviewed with such a holistic approach. There was little relationship between order policies for the parent items and their components. One of the teams discovered that a body lotion was manufactured in bulk batches of 5,000 liters, but the filling and packaging lot size was set up for the equivalent of 6,000 liters. As a result, to package a batch of product, two lots of bulk product were produced with the surplus held in inventory. A good deal of that surplus routinely expired and was scrapped before the next packaging run. The same team then began to find errors in component quantities when comparing the system quantities against the product engineering design documents. The team reconvened at 5:00 P.M to finish its assigned products. Late that evening, Francis Brandon entered all the corrected master data into the ERP system. The team had finished its first task, but realized there was far more work to do than they had anticipated. And they hadn’t even looked at Work Location or Routing records.

Process Design team leaders met with their teams to review and reach agreement on the approved output-Gantt chart. After some initial push-back to try to get more time and to argue about dates, they got the message. The dates were firm and would be changed only if there were a lack of resources. Greg’s support of this initiative ensured that they would always find the necessary resources. Peter had to work directly with three managers who were reluctant to release their team representatives.

At the same time, the Capable Integrated Business Management teams participated in the five workshops for the Integrated Business Management process [see Figure 6.2]. Within a period of 10 days, they had designed the process and procedures for all the steps and established a calendar, ambitiously starting the very next month. They realized that all the required input data would not yet be complete at this stage, but knew that everyone would benefit by following the process and learning to work as a team. With such a short time line, they would use current data in the new format. They would find gaps in the data but would use what they learned to improve overall quality of the data. As data improved, so would the process.

The teams learned quickly and improved their procedures. Spreadsheets were created and current master data downloaded from the ERP system to facilitate review, correction, and automatic uploading back into the Test database to perform simulations. Jeff Black quickly developed representative demand plans for the 5 SKUs and sent them to Janice Hackworth, who began developing the Master Supply Plan. Janice and her Supply Planning team quickly learned that no one understood the software module well enough to create plans in the system. She used spreadsheets for now to accelerate the process design work, and began scheduling planners for necessary system retraining. Procedures were developed to allow MRP to be updated several times each day if necessary. Notes were kept by each team so that they could later formally document the new policies and procedures they were developing.

All the teams were staying on schedule and meeting due dates for Planning Spine development. The Cosmetics-Test database was being loaded and the test script was nearing completion.

It was time to start the test. The first run would represent an “ideal” state, assuming no problems. Janice assigned the work stations and the various supply planning roles. The first script covered three days; IT would provide an historical record of the test so that the same scenario could repeated later but with specific problems, such as a production batch failure, introduced.

The team was nervous but excited as they assembled for the first three-day scenario. Peter reminded them that one of their tasks was to discover and correct errors but the primary job was to make the model simulate their business. Until they achieved that objective, they could not move on. Peter and Janice would move among the workstations to observe and coach. They executed the first MRP run in 20 minutes. The date was advanced to “Day-2,” and the master supply plan was “exploded” through all levels to create an end-to-end supply plan. Action message reports were divided and distributed by planner code. Each planner reviewed the messages to see what they could learn about their plan. If the master data was correct and loaded perfectly in the system, there would be no error messages. That’s not what happened.

Of the 183 action messages in the report, only about 80 were expected. The team spent three hours tracking the sources of the errors, and determining the root causes for elimination and learning. Many were caused by inventory imbalances within the plan that caused reschedule in or out recommendations. These problems were corrected quickly. There were only a few BOMs, but inventory imbalances at the top level cascaded down through all planning levels causing multiple reschedule messages. After they completed the corrections, Peter ran the MRP scenario again. This time there were no messages. Now they were ready to begin planning!

Bob Malinkov and Chris Deutsh from IT set up a “lessons learned” archive in the database so that issues could be numbered and tracked until they were closed. They also set up a projector so that workstations could be projected to facilitate learning for the entire team. First they projected a demonstration of releasing orders level by level to see how the system worked. Peter was observing Richard Jones working with the bulk manufacturing simulation and motioned for Janice to join him. Richard had uploaded data from his familiar planning spreadsheet and was manually developing a manufacturing schedule. Janice asked him to project his plan and got everyone’s attention. “Richard, please explain what you are doing.”

“Sure, Janice. I noticed that the order dates and quantities didn’t look right, so I . . .”

Janice interrupted. “When did you notice they were wrong?”

“When I was watching the first simulation, I spotted the problem and knew I could fix the problem easily in my spreadsheet. The spreadsheet is easy to use and has worked really well since I developed it 18 months ago.”

Janice looked at the other team members, “Anything wrong with what Richard did?”

The other planners were quiet. Most of them would have done exactly the same thing if they hadn’t been running this scenario. Bill Bates from Production Operations finally spoke up.

“If I’ve understood it right, we should not need any spreadsheets to support MRP. I think that’s what Janice meant when she said we have to make the planning model work inside the system. So if there’s anything wrong, Richard, you should have stopped the scenario so we could all learn how to find and fix the root cause.”

Peter jumped in, “That’s exactly what we want to do. No spreadsheets to run the scenario and the business. Find out what’s wrong and get it cleaned up in the system. Roxanne told me that the planners face the toughest challenge. They have to unlearn the way they’ve been doing their work. Now I know what she meant! So thanks for the learning opportunity, Richard.”

Richard responded. “You’re welcome, Peter. I’m learning fast, but not fast enough. I know I have to get my other work done, but I need to work on this more often than in our weekly meetings. Is it possible for the team to come to the Customer Service conference room every morning from say 7:00 A.M. to 9:00 A.M. to work on the model until we’ve got our Planning Spine nailed down? I think that would accelerate our progress.” Peter was pleased with this idea, and all the team members agreed.

As the team worked through more difficult scenarios to build more detail in the Planning Spine, they began documenting their new processes. They were developing confidence in their processes, in the system, and in their ability to get rid of the spreadsheets they had depended on to run the business.

Over the next few days, the team “lived” through 15 MRP days in the model, basing all order planning on recommendations from MRP. They started to build a Planning Manual containing the new Cosmetics Planning and Control procedures and standardized approaches to planning and problem solving. They were also developing flow diagrams to help everyone understand how the processes were integrated and to clarify inputs, outputs, and accountabilities for executing the process. There would be a copy of this manual on every planner’s desk. Along with the CBT modules being developed, the manual would be required reading for all new planners.

Simulations increased in difficulty to include the kind of major problems that the planners faced on very bad days. Confidence continued to build as the team learned how the planning system could help them cope with these problems and avoid customer service failures. A few teams completed their tasks, but most came across difficult design problems. Their Effective Management coaches had seen many of the problems before and were helpful in guiding the teams in the right direction through coaching and additional detailed education.

The Demand Management process design team, led by Jeff Black, received a great deal of early management attention. The current demand planners were confident that their business was not forecastable. Education began with a two-day Managing Demand course that Roxanne tailored to the Cosmetics Products business. It would be presented by EM’s fashion industry forecasting expert, Liam Lawlor, who had been a Demand Manager in both fashion and catalog companies. Peter and David attended, as did Alexandra, but only for the first day because of an important customer meeting. Liam reviewed the classic statistically based forecasting model, very close to what the demand planners used. He next showed them a slide of quotes from many companies [Figure 15.4].

Everyone laughed as they recognized that they, too, had made similar comments about forecasting, except for the last. He asked them if they thought supply could produce better forecasts than sales or marketing, and if so, why. Liam turned to Janice Hackworth, who was sitting in on the class to learn more about her interfaces with the Demand Planning Manager.

“Any ideas, Janice?”

“Frankly, we have to second-guess the forecast because the SKU forecasts we receive from Sales and Marketing are so inaccurate. The only thing that comes to mind about our forecasting process is that we don’t even try to plan at SKU level in supply except for the next few days in packaging. We tend to plan at the bulk product level.”

“If I can restate what Janice said, supply doesn’t forecast at SKU level. Why not, Janice?”

“Because the SKU demand mix changes all the time. We have long cumulative lead times and so have to plan at a more aggregated level where historical demand is more consistent over time. The packaging lines and packaging materials have shorter lead times than the bulk materials. So, we can more readily change packaging schedules to deal with mix changes.”

Liam then demonstrated that their current forecasting model was unrealistic for their needs, and indeed would never produce effective forecasts.

“I need to remind you that Supply shouldn’t be forecasting at all. Their job is to create and execute plans that support the forecast. Have you considered you may be forecasting demand at the wrong level? That some of your SKUs are very forecastable, but most of your smaller SKU’s might be unforecastable?”

He also suggested that they might be too focused on the numbers, and explained that effective forecasting is built on assumptions that are the foundation for the calculated numbers. Assumptions, he said, must be explicit, documented, and reviewed every month. Focusing on the assumptions rather than the resulting numbers would quickly improve understanding of markets, and produce better numbers. “I won’t take you through this example yet, but I want to show you the assumption format that one of my clients uses. It is simple, but effective [Figure 15.5]. You’ll see more of this in the Demand Planning course.”

Liam next described unconstrained demand—a forecast unconstrained by any consideration of supply constraints. The forecasters liked this concept since they often felt schizophrenic trying to create forecasts they knew could not be supported by Supply. Out of self-defense they would alter the true forecast to match what they thought could be produced. Liam explained that, when necessary, the forecast would be constrained to match supply, but later in the Integrated Business Management process through Integrated Reconciliation.

In response to a complaint about their statistical forecasting tool unreliability, Liam discovered that the forecasters didn’t really understand the ability to modify the forecasting algorithm. “This isn’t a subject for the Demand Planning course, but I want to point out that you do have a great deal of control with that tool, and with the overall forecasting process, that you don’t seem to be exploiting.” He explained a number of forecasting algorithms used by most tools: time series, simulation, probability, Delphic for completely new products and, ultimately, Collaborative Demand Planning involving key customers. The team was listening and visibly overwhelmed by these new concepts.

“Don’t worry yet about how you will select the best algorithm. Just be aware of this important principle. Whatever algorithm or model or process you use, you must always test the forecast bottom-up versus top-down using the same set of assumptions. By that, I mean aggregate your detailed forecasts up to category and family forecasts. Then look at the family and category history and disaggregate them down applying assumptions for the future and compare the outcome of the two processes. Call it a ‘sanity check’ if you will. Reconcile any significant differences and modify your assumptions where necessary. I can tell you that your forecasts will begin to improve immediately. I can also tell you that your business is absolutely forecastable!” Liam now turned to Jeff Black.

“Now Jeff, let me ask you a fundamental question about forecasting. What is the greatest source of forecast error?”

Jeff offered his opinion. “Based on what you just said and our experience, I suppose it is a pretty close tie between not knowing our customers well enough and using the wrong forecasting method.”

“Those are certainly contributors, Jeff, but not the biggest cause for most businesses. The main contributor to forecast error is . . . long manufacturing lead times. The longer the cumulative lead time, the greater the exposure to forecast error! I want you all to consider this [Figure 15.6].

“If your lead time for end-item sales were one week, you’d probably be in a Finish-to-Order (FTO) supply model; you would have pretty good SKU forecasts, wouldn’t you?”

Janice replied, “Well, yes. But I don’t see how that could work for us. Our packaging lead time alone is two weeks, not counting some long manufacturing and supplier lead times. That’s why we have to have SKU forecasts at three months, and a detailed review of them at one month before we get into actual production.”

“I understand, Janice. With your current process, that makes sense. But suppose you asked Sales and Marketing for forecasts at the product level, not the SKU level. That more aggregated forecast would be more reliable. The law of big numbers says that variability within a larger population is normally less than variability in a smaller population. The forecast still won’t be perfect, but you would experience less forecast error. Isn’t it the variability in the small SKUs that causes you the greatest problems?”

“Liam, that makes a lot of sense, but we make to stock. We’re not set up to make the products after we receive orders.”

“Good point, but now is the perfect time to consider what it would take to change your supply model as you design your Planning Spine. What would it take to create your Master Supply Plan at the product level; and finish the SKUs based on customer and distribution orders? Companies in your industry who make the change achieve significant bottom-line benefits. They delay committing bulk product to final packages until the last minute and see savings through massive reductions in SKU inventory, damage, obsolescence, and warehouse costs.” Janice thought for a moment before responding.

“I don’t know what it will take but it sounds intriguing. We can certainly use the working capital cash that would be freed up and the reduced inventory! Peter, we had better take that as an action item for the Planning and Control Milestone team. Let’s propose we develop a FTO model and see what it would take for us to get there.”

Liam added a point of clarification. “Peter, the model won’t have to be either/or. In all likelihood you’ll end up with a combination of FTO and MTS, that’s Make-to-Stock. For some of your small volume SKUs, you probably won’t want to incur frequent setup changes and will simply make to a target warehouse inventory. It’s a trade-off. As a first step, you might consider getting your top ten or so SKUs that you are set up to run every week or day, or so, into an FTO model. That will probably give you most of the benefits.

“Finally I’d like to leave you with this summary—for both your milestones [Figure 15.7].”

Two early results were reported to Peter for the weekly Steering Committee meeting. The Inventory Accuracy Team, following the multistep methodology taught by Effective Management, had determined their baseline accuracy. They were shocked to discover that their inventory records were only 43 percent accurate. This reinforced the reasons why planners were reluctant to give up their spreadsheets and rely on ERP action messages instead. The team reported that they were moving aggressively to implement control group cycle counting to discover and eliminate the root causes of inaccuracies.

The Data Integrity Team, similarly, had completed its baseline audit of Bill of Material accuracy. They sampled more bills than they planned because accuracy was far worse than expected. To their astonishment, baseline accuracy was 18 percent. They found problems with system access security, change control, scrap, and yield factors—additional reinforcement of the underlying reluctance to use the ERP system. They recorded the results of their investigation in the eList. They looked forward to their scheduled tasks with trepidation, but also with excitement that the work would contribute directly to improved customer service.

Roxanne assured Peter that these inventory and data accuracy results were not unusual at the start. She also assured him that implementing best practice business processes would enable Cosmetics Products to improve use of the ERP system, planning, and as a result, customer service. Peter wasn’t looking forward to reporting these dismal accuracy results to the Steering Committee.

At the end of the Steering Committee meeting, Greg asked Peter to thank the teams for their progress and reports. He, too, was a bit shocked by the accuracy baselines, but he was also encouraged. “Look,” Greg said to Peter, “the accuracy is no worse than it was yesterday. The good news is that today, we know exactly where we stand. I’m encouraged that the teams now know the gaps, the root causes, and have plans in place to reach at least 95 percent.”

He concluded the Steering Committee meeting by saying, “I’ve never been more confident that we can meet our customer service objective. Keep up the progress. The train I mentioned in the kick-off meeting is definitely moving!”