Toyota Sales and Operations Planning

Sales and operations planning (S&OP) is a critical component of the supply chain planning process. It is linked upstream to the mix planning process and downstream to the production scheduling process. The goal of S&OP is to generate a production plan that balances demand and supply in a profitable way. The end point of this process is an order for vehicles with full specifications that will be scheduled for production (see Production Scheduling and Operations).

To understand the Toyota approach to sales and operations planning, an understanding of which entity in the supply chain submits the vehicle order is necessary. Simply put, it is not the dealers. At Toyota, the regional offices submit the vehicle orders once each month. Toyota uses a monthly allocation process to allocate the vehicles planned for production to each dealer—in other words, it uses a top-down approach. Many other automobile companies create production plans from the bottom up; that is, they collect the orders submitted by dealers and then create a production plan. There are pros and cons to both approaches; however, the top-down approach works well for Toyota because it enables Toyota to ensure that the resulting production plan is stable.

Sales and operations planning processes at Toyota are performed in two stages: annual planning and monthly ordering. Annual Planning The objective of the annual planning process is to establish a rolling three-year sales and production forecast. The process is repeated semiannually so that the forecast is updated based on the latest market and economic conditions. The annual forecast is used throughout the company to project profits, establish capital and operating budgets, evaluate plant and supplier capacity requirements, conduct annual price reviews with suppliers, and influence marketing strategies.

Annual Planning

Process The annual planning process is a collaborative process between the sales and manufacturing divisions. The responsibilities of sales are to grasp the market and economic conditions, predict competitive product plans and strategies, and understand new product launches and marketing plans to create a sales forecast for each model for each month and year. Manufacturing’s responsibilities are to determine the operating capacity for each model and each plant, evaluate various model mix scenarios, and identify peaks and valleys in the production calendar that are created by model changeover schedules.

The process is executed over a period of several weeks twice a year. There is a push-pull momentum: sales submits its request for each model by month and year, while manufacturing simultaneously attempts to mix and match the model volumes across the plants. Collaboration is required because typically the sales request is different from manufacturing’s capabilities and/or objectives.

Typical sales objectives are to remain flexible to respond quickly to market changes and to limit the use of incentives to sell vehicles. Some of manufacturing’s key objectives are to operate at full capacity and to produce high-profit vehicles. Because most plants produce multiple models and many models are produced at multiple plants, numerous scenarios must be considered. Some significant conflicts may need to be resolved, which results in the push-pull momentum, or the give-and-take exchange, between sales and manufacturing. The collaboration process focuses on adjusting the variables that can be used to bridge the gap. Some of these adjustments to the sales and operations plan are as follows:

  • Sales can plan to use incentives selectively to create demand for slowselling models and for models that are scheduled for a major model changeover.
  • Sales can adjust marketing strategies to promote selected models. For example, special-edition models may be created to enhance a model’s marketability. This strategy is used mostly during the later years of a model’s life. (Note: Most automobile models are produced for about five years before there is a major model change.)
  • Manufacturing can upgrade facilities to increase capacity for selected models.
  • Manufacturing can adjust the speed or takt time of the assembly line at a plant to increase or decrease production.
  • Manufacturing can vary the model mix within a plant.
  • Manufacturing can vary the planned overtime by plant (e.g., schedule overtime on Saturdays).

One example of collaboration between sales and manufacturing that generated significant benefits occurred at the Toyota plant located in Fremont, California. The California plant could produce only 20 percent of the Corolla models with sunroofs. The Corolla model was also produced at another Toyota plant near Toronto, Canada, that did not have this constraint.

The demand from the West Coast for Corollas with sunroofs was very high. Unfortunately, the California plant could not produce enough vehicles with sunroofs to meet the demand, so some of those vehicles destined for the West Coast had to be produced at the plant in Canada. The result was an extremely negative impact on vehicle distribution. The vehicle distribution group preferred to ship the Corollas produced in California to dealers west of the Mississippi and the Corollas produced in Canada to the dealers east of the Mississippi. The reasons are obvious: logistics costs are lower and the time to get the cars from the plant to the dealers is shorter because of the geographic proximity of dealer locations to the assembly plant.

The dealers located in the West were requesting over 40 percent of the Corollas equipped with sunroofs. The consequence of the inability of the California plant to produce Corollas with sunroofs to meet this high demand was that many cars produced in Canada were shipped across the country to the West Coast and an equal number of Corollas without sunroofs were built in California and shipped back East. At each annual planning meeting, people from the sales group would argue that this situation was unacceptable and request the California plant to install additional equipment to eliminate this constraint. The plant management was very cost conscious and would not agree to invest the millions of dollars required to install the equipment. The stalemate continued for several years until there was a detailed analysis done to calculate the profit impact on both sales and the plant. It was determined that Corollas that were ordered with sunroofs normally were equipped with other higher-profit options. Thus, the profit margin of the Corollas without sunroofs was significantly less than that of those with sunroofs. It was determined that by increasing the mix of cars with sunroofs built in California from 20 percent to 40 percent, to respond to sales demand, the plant would recover its multi-million-dollar investment in a few months. When the plant manager realized this missed profit opportunity, he quickly agreed to install the necessary equipment.

Another example of the types of decisions that are made during the annual planning process is found in a July 2008 Toyota press release:

July 10, 2008—Erlanger, KY—Toyota is responding to changes in consumer demand and improving the production efficiency and stability of its North American operations by adjusting production mix at three plants. The changes include the addition of the Prius hybrid sedan to its North American lineup. The changes are as follows:

  • The Prius will be built at a plant under construction in Blue Springs, Miss. Production is scheduled to begin in late 2010. Prius, which will join the Kentucky-built Camry Hybrid as the second Toyota hybrid built in North America, enables Toyota to better respond to increased consumer demand for hybrid vehicles.
  • The Highlander mid-size SUV, originally scheduled to be built in Mississippi, will now be manufactured in Princeton, Ind., beginning in fall 2009.
  • Production of the Tundra full-size pick-up truck, currently built in Indiana and Texas, will be consolidated at the San Antonio plant in spring 2009.

This is an excellent example of how Toyota has the agility to make major changes in its production schedules in a relatively short time frame.

Component Planning

After several weeks of discussions and analysis, the annual plan is finalized by model by plant. Then the focus shifts to determining the ratio of key components of vehicles. For example, engines, transmissions, and similar options may have a capacity constraint at the supplier level. One of the complications of planning component volumes is that most components are used by multiple models across multiple plants worldwide. Again, there is a need to run several scenarios to determine how to best balance component plant production plans with the various assembly plants. As you can imagine, this is like trying to solve a giant Rubik’s Cube.

Creating Buy-In

Another important point is that the sales and operations planning process cannot be resolved simply by inputting variables into a computer and getting a mathematically accurate result. Certainly, computers play an important role in calculating the various scenarios; however, the human interaction that takes place over the hours, days, and weeks not only improves the quality of the annual plan but also builds an overall consensus among sales and manufacturing groups that reinforces teamwork. This creates an environment to achieve success by motivating all parties to work together to ensure that the annual plan is a commitment for both the sales and manufacturing organizations. Such collaboration is not the normal process in different functional disciplines. They tend to stick to their viewpoint—not because they somehow benefit from doing so, but because they do not see the viewpoint of others. In such cases, there is much more reliance on computer results than on human interaction. So Toyota prefers a combination approach. That is part of the Toyota Way, in which the human-machine combination is often assumed to produce the best possible result.

It is interesting to observe how many Toyota Way principles are demonstrated during the annual planning process. Teamwork and mutual respect are key principles used throughout the planning process. Moreover, the process through which the annual plan is reviewed and updated semiannually is a good example of how Toyota practices one of the key Toyota Way principles: Plan, Do, Check, and Act (PDCA). The plan is developed and is used as a basis for day-to-day operations. Then it is checked after six months, and action is taken to adjust the plan and/or make changes in plant or supplier capacity. This results in a closed-loop control that is repeated over and over again. PDCA is also applied to the planning process itself.

After each annual planning cycle, a reflection report is prepared to evaluate the process. This report highlights what went right and what needs improvement. The items that need improvement are analyzed further to identify root causes and countermeasures that should be implemented.

Capacity Planning

Capacity planning consists of two aspects: the internal plant capacity review and the supplier capacity review.

Internal Capacity Planning

Each assembly plant conducts a periodic review of its production capacity. The purpose is to calculate the upper and lower limits of its operations rate for each product. The capacity planning activities are performed on an annual (or, in some cases, a semiannual) basis. However, most major investments to the facilities are usually implemented during the next major model change. For assembly plants, the products are the various vehicle models that are produced on each assembly line. For component or unit plants, the products are the components such as engines and transmissions. Using the annual plan volumes as a forecast, each production facility will consider the variables that impact capacity. A key constraint is how much the operations rate varies by assembly plant. The following are some of the variables that impact the operations rate and thus can be adjusted:

  • Direct labor and the flexibility to add workers—or reduce the workforce— and thus, adjust the production rate. Because Toyota strives to avoid layoffs of Toyota employees, most Toyota plants employ some percentage of temporary workers to support the normal production level. If production needs to slow down because of slow demand, then temporary workers can be reduced. On the other hand, if production needs to be increased, then, most likely, temporary workers would be added initially. If the increase appears to be permanent, then some temporary workers would be converted to full-time employees.
  • Facility and equipment. Each process must be analyzed to determine the weakest link in the production process. In other words, even if one could add unlimited numbers of workers, there will be some equipment that would not be able to produce at a higher rate. It could be a machine that is used to install a sunroof, perhaps an additional paint booth might need to be installed, and so on.

Supplier Capacity

So that potential weak links in the supply chain can be identified, supplier capacity as well as internal capabilities must be evaluated. Because this is a joint responsibility, purchasing and production control need to work with each supplier to identify any constraints that could restrict the supply of a part. Although most companies focus attention on the upper limit of each supplier’s capacity, it is also important for a supplier to highlight any planned production cutback that would severely impact its ability to operate. Thus, a result of this capacity planning study is for Toyota to document the upper and lower range of production for each part and supplier.

Proactive steps can be initiated to solve the capacity issue. For example, a supplier could be added if production volume for a part were increasing rapidly. Such a step would provide Toyota with a backup situation in the event its primary supplier has a problem. But there are other reasons that Toyota would consider developing a dual source for some parts. Some of these considerations are risks because of location of suppliers that could restrict supply (e.g., poor weather or the potential for flooding or earthquakes) or the financial stability of the supplier. Thus capacity planning provides a perspective of overall production rate that can be executed across the supply chain.

Monthly Order

At Toyota there is a monthly global process to receive the sales orders from each sales company from around the world. That is translated into a production plan for each assembly plant as well as for each Toyota unit plant. Toyota’s culture emphasizes a process that does not rely only on sophisticated computer systems. Although Toyota certainly utilizes numerous computer systems that process data and crunch the numbers, the results provided by the computers are reviewed and discussed by a cross-functional team of sales and manufacturing managers. The process is an iterative one that ultimately generates a three-month rolling production plan for all Toyota assembly and unit plants worldwide. A joint focus by both sales and manufacturing on the monthly order ensures that all perspectives are balanced and the logic for the decision is clarified.

Production Calendar

One prerequisite to creating the production plan is to determine the number of production workdays in a production month. One of the reasons that a production month is not the same as a calendar month is because some of the processes within manufacturing are based on a weekly cycle. The concept of a production month was implemented so that each month would consist of either four or five complete weeks. Each year, a production calendar is created based on the following rules:

  1. Each week is assigned a sequential number from 1 to 53.
  2. Week number 1 is always the week that includes the first production day of the year. For example, if January 1 is on Wednesday and January 2 is the first workday, then the week will be designated production week 1. Note: In this case, this week will actually include two days of December. Although this breaks the next rule, it is not considered an issue because in the auto industry in the United States the last week of the year is a nonproduction week.
  3. The month into which each Monday falls is used to determine which weeks are classified into a production month except for January. For example, if April 30 is a Monday, then the entire week is included in the April production month.

Because there are 52.2 weeks in a normal 365-day year, there will be some years that contain 53 weeks. The production calendar is published prior to the beginning of the calendar year and includes the production week classification as well as company holidays observed by each plant. Use of a production calendar prevents confusion regarding planning across calendar months and preserves the weekly planning process.

Production Planning Process

Once the production calendar is published, it can be used to create a monthly production order for each plant. The monthly order process is a global process managed by Toyota Motor Corporation (TMC) in Japan. Each manufacturing assembly plant creates a preliminary operations plan that shows the quantity of each model that will be produced, including critical options such as engines and transmissions. In addition, each unit plant creates an operations plan that shows how many engines, transmissions, and other options it can produce.

An example of an operations plan is shown in Sample Operations Plan. In this example, one week of a month is shown; however, the complete plan would show all weeks in each of the three months that make up the rolling three-month plan. The key components of this plan are as follows:

  • Standard work hours. In this example, Monday is a holiday, so work hours are zero. For all other days, standard work hours are 16, because most plants work two 8-hour production shifts.
  • Overtime. Normally there is some limited amount of overtime that is prescheduled. In this example, it is two hours (or one hour per shift) per day. Day-to-day adjustments to the overtime are made just prior to the start of each shift based on operational conditions at the plant.
  • Takt time. Takt time is a term that refers to the speed of the assembly line. This unit of measure is the amount of time it takes a vehicle to move from one station to the next. In this example, a takt time of 60 seconds means that workers assigned to each station along the line have 60 seconds to complete their work. The rate of vehicles completed and driven off the line is one per minute or one every 60 seconds.
  • Operations rate. The percentage of the time that the assembly line is running at the normal production rate is the operations rate. That is usually set at less than 100 percent in a Toyota plant, because Toyota emphasizes quality first. So all workers are encouraged to stop the line if any problem occurs so that problems can be corrected before the vehicle gets produced. That measure also ensures that problems that may affect multiple vehicles are identified and corrected early. The operations rate is reduced by the time required to fix the problem. Thus, if the takt time is 60 seconds, 1 percent of the production is interrupted by a line stoppage, and it takes 5 minutes (300 seconds) to fix the problem, then the expected operations rate 60/(60 0.01 300) 60/63 95.2%.
  • Vehicles per hour. This is calculated as seconds in an hour (360) divided by takt time in seconds (60) times the operations rate.
  • Vehicles per day. This is calculated as total work hours per day times the number of vehicles per hour.
  • Model ratio. Most Toyota assembly plants produce multiple vehicle models. Therefore, the ratio of each model as a percent of the total of all models must be determined to create a production volume by model.
  • Model volume. This is calculated as model ratio times the production volume.

As you can see from the sample operations plan , all of the components are variables that can be manipulated to create what-if scenarios that can be evaluated during the negotiation process. These metrics, or key performance indicators (KPIs), are also used on a day-to-day basis to monitor the plant operations.

The monthly planning process requires input from all sales and manufacturing operations worldwide. Each sales company submits its request for production of each model by month. This information is transmitted to TMC in a file that contains a record for each vehicle, including full specifications. Although the sales request contains the full vehicle specifications, this initial planning step considers volumes only by model and key components such as engines. The total worldwide demand is compared to the proposed operations plan for each plant. Worldwide Sales Request shows a sample of how vehicle and engine volumes by month are summarized by the global regions.

The next step is to balance the sales request with the production operations proposal to determine the optimum sales and operations plan. Then the production volume and model mix for each plant is allocated to each of the global sales regions.

Global/Regional Allocation

Each of the sales companies receives a share of the production for each model and each assembly plant from TMC for each of the rolling three months. Thus, each sales company (for example, Toyota Motor Sales, USA) must submit a monthly order and two months of forecast. The order month is designated as “N.” The first forecast month is “N 1” and the second is “N 2.” So, if July is the order month “N,” then August is “N 1,” and September is “N 2.” Some references classify these months as PPR1, PPR2, and PPR3, where PPR stands for “production plan requirements.”

Each sales company then subdivides its volume into an allocation for its regions and/or countries. For example, in the United States, the allocation is broken down into 12 regions, and in Europe it is allocated across more than 25 countries. The volume by region by model will vary based on performance versus sales objectives for each region.

Next we will examine how each region can manipulate the content of its allocated vehicles to match its regional mix.

Final Sales Order Preparation

The headquarters for each sales company will aggregate the order of each of its regions and transmit that information to TMC. During the aggregation process, the sales headquarters must do a final check to see to it that the total order complies with TMC’s guidelines.

The following paragraphs show a scenario that a region would follow to generate its monthly order and forecast. (A sample of only one region’s data is used, for purposes of keeping the illustration simple.) The first step is to analyze the current stock situation in the region. That is done by counting all stock by build combination as well as by color. (In this scenario, color is used as an example; however, the same process is done for each build combination.) Stock is classified as dealer stock or pipeline stock: dealer stock is owned by the dealer and either is physically at the dealer or will arrive within two to three days; pipeline stock is further divided into in-transit and allocated stock. In-transit stock is completed vehicles that are at the plant and/or in-transit to the dealers. Allocated stock is production that has been scheduled for a region but has not been built.

For this example, we have assumed that the quantity of each of these categories is equal to one month. Obviously, that figure will vary depending on many factors such as recent demand, in-transit time, and allocation point.

Stock Analysis how stock is calculated across the pipeline by each build combination and color. Then the mix of stock is calculated. This example is just showing colors; however, these processes need to be repeated for each build combination and color across all models.

Next the stock mix is compared to the target mix. As discussed in Mix Planning, the target mix is determined prior to new model introduction and may be adjusted periodically if market conditions change. The variance of the actual stock quantity compared to the quantity computed using the target mix ratio is used to make an order adjustment. Those calculations are shown in Mix Analysis.

Next the three-month order and forecast is calculated based on the target mix

Finally, the order adjustment is used to adjust each month’s quantity to rebalance the actual total stock to equal the target mix. Note that the adjustment is applied over a three-month period to avoid overreacting to changes in demand. Stock Analysis . Mix Analysis . However, if the regional manager has some additional information that would indicate a more sudden change, he or she may make a decision to override this calculation.

These processes are then repeated for each package or build combination. As was discussed in Mix Planning, each region will limit the number of build combinations ordered for stock to the high-volume sellers. Hhow the total number of units may be distributed across the four packages for the three-month period.

The order can now be completed by combining the color quantities with the quantities for each package. That step is necessary because manufacturing requires the complete vehicle specification to determine the production schedule and eventually the part orders. Shown in the final composition of the orders. Each of the four packages is ordered with five different colors.

The process just described assumes that there are no unusual conditions that would affect the regional order. Although such normal circumstances may occur on some models for some months, most of the time external conditions will require that the order be modified. Some of these conditions are special fleet orders, seasonal trends, and special sales promotions.

Special Conditions

Fleet orders can occur at two levels. The national fleet accounts are managed by the sales headquarters. Each month, allocation to all of the regions is reduced by the amount of the national fleet orders. The fleet order is then submitted by the fleet department. In addition to these national fleet orders, each region may have local fleet orders. So the region would subtract the number of vehicles from its allocated amount and create special orders based on the unique fleet requirements.

Seasonal trends can also create a need to modify the mix of the order. Therefore, the targeted mix would change in a stair step manner to reflect projected seasonal changes. For example, convertibles and sunroofs sell better in spring and summer months. Also, lighter colors are preferred in hotter weather.

Special sales promotions also can require that the order be modified. For example, air-conditioning may be included for free if you purchase a car with a sunroof. Obviously, that would require the normal mix calculation to be overridden to ensure that an adequate supply of cars with sunroofs and airconditioning is ordered.

The point is that a computer system can be programmed to calculate the monthly order if there are no external factors. However, the process requires human intervention to evaluate the computer-generated order and make necessary adjustments. That usually requires multiple iterations before the order is finalized.

A computer making trade-offs might ignore some subtle constraints such as option combinations that are questionable. Human interaction certainly could have helped in the case reported in the Economist on January 31, 2002 (story told by Hau Lee): “In the mid-1990s, [Volvo] the Swedish car manufacturer found itself with excessive stocks of green cars. To move them along, the sales and marketing departments began offering attractive special deals, so green cars started to sell. But nobody had told the manufacturing department about the promotions. It noted the increase in sales, read it as a sign that consumers had started to like green, and ramped up production.”

Sales Aggregation and Adjustment

Once each region has created its order, it is transmitted to the sales headquarters where it is aggregated to produce a national order to be sent to TMC. The order is checked to ensure that each region has ordered the correct number for each model. Next the order contents must be checked against the manufacturing constraints to ensure that there are not any constraints that are exceeded. For example, there are several controlled specifications such as engines, transmissions, and wheels that may have upper and lower limits. These limits are based on capacity restrictions at Toyota unit plants and key suppliers. In the event that one or more constraints are exceeded, the order can be modified by prorating the adjustment across all regions or, if necessary, manually making adjustments to selected regions.

Keep in mind that even though each region order consists of a package and a color mix that sells best in its particular region, the national aggregate mix will normally be more evenly distributed. In addition, any change from month to month will be muted. How a very different mix of build combinations by sample of four regions will, when aggregated, create a more even mix of build combinations at a national level.

Once the sales order is completed, it is forwarded onto TMC in Japan. TMC aggregates all orders from all sales companies worldwide. Then the sales requests are compared with the production operations plans submitted by each manufacturing plant. High-level meetings attended by sales and production people are held to resolve the differences between sales requests and production operations plans. The result is that some last-minute adjustments will be made to ensure that the optimum production order is submitted. That collaboration process is similar to the annual planning process, with one big exception: the time frame to complete the negotiations is extremely tight. During the annual planning process, the elapsed time to resolve differences is measured in days or weeks, whereas during the monthly process the order must be finalized within one or two days.

One metric to measure accuracy of the forecast is to compare the final order to the previous forecast and measure percent of change or forecast accuracy for each option. If there are any extreme changes, they should be investigated because they may be an indication of an order error.

How Does Toyota’s Method Compare to Other Planning Methods?

In many firms, the sales and operations planning process focuses on materials planning constrained by capacity. Detailed accounts of different processes adopted by firms can be obtained from standard books, such as that of Vollmann, Berry, Whybark, and Jacobs.2 In manufacturing planning and control literature, the coordination between sales and production takes place via the “production planning” process. In addition, some firms might use distribution or sales requirements planning to send shipments from the plant to different regions or warehouses. As Vollmann and others point out, the modern term for such coordination is “sales and operations planning (S&OP).” That process begins with a sales forecast for some predetermined horizon, say 12 to 18 months. Increase and decrease in inventory levels as well as resources are planned during this process. Limits are placed on the amount of possible increase within, say, three months and for the period three months and beyond. Typically, these plans are reviewed and revised on a monthly basis. In traditional production planning literature, this step is also called “aggregate production planning.” The plans are made for groups of similar products, called families, therefore said to be aggregated. The costs considered in production planning are the costs of carrying inventory and changing workforce level, as well as the cost of overtime. As with the Toyota process, the goal is to determine overall sales rates, production rates, aggregate inventories, and backlogs.

As can be discerned from the description, a significant amount of coordination needs to be carried out through meetings. In fact, many writers have emphasized the need for organizational change for successful implementation of S&OP. Therefore, in recent years, the lack of coordination has been explicitly addressed by the introduction of a high-level sales and operations planning process called the “executive S&OP.” SAP,3 for example, offers two blends of S&OP: SAP SOP for executive S&OP and SCM for near-term S&OP. Moreover, the S&OP is said to be the business process for coordinating supply, demand, and financial plans. As Wallace and Stahl4 have stated, “The term ‘Sales & Operations Planning’ traditionally referred to a decision-making process for balancing demand and supply in aggregate. This is an executive-centered activity.” They go on to say that in recent years the term S&OP has been broadened to include detailed planning at the product and customer order level. It is also somewhat clear from the description given previously that the planning process in other firms is not constrained by “overall” guidelines beyond profit maximization or cost minimization.

In Toyota, these “overall” guidelines also control the overall process, namely, paying attention to stability, careful planning before reacting, and measurement of forecast accuracy and consensus building before making changes. These guidelines emphasize the need to consider the impact beyond the immediate functional or firm boundary. In our opinion, such explicit guidelines differentiate the S&OP process at Toyota from that of other firms. Moreover, the performance of the plan and the planning process itself are measured against carefully selected metrics; they are not left to vague or conflicting interpretation.

Reflection Points

Can a firm that does not mass-produce adopt the Toyota S&OP principles? The main principles adopted by Toyota are straightforward and summarized under the v4L framework:

  • Variety of cars sold is managed by keeping the national aggregate mix of packages and colors stable across time.
  • Velocity of sales in a region is adjusted to match feasible production rates (takt time) at plants. Capacity planning estimates define the upper and lower production rates across the supply chain.
  • Variability of production is managed by freezing sales commitments over specified periods. Variability of sales is managed by adjusting sales incentives to deliver planned sales.
  • Visibility is ensured by tightly linking sales and operations plans, developing them collaboratively with buy-in, and deploying them across the system.

The key learning principles used are as follows:

  • Create awareness. Deviations from plans are made immediately evident by the use of key metrics. For example, if one of the constraints is exceeded, that fact is highlighted by one of the computer checks and forwarded to a planner to investigate.
  • Make action protocols. Methods for taking corrective action are clearly documented. For example, in case of a mismatch between operations and sales plan, the sequence of actions is to first clearly identify the gap and then to consider the potential countermeasures, gain consensus on corrective action, and implement changes.
  • Generate system-level awareness. Systemwide considerations are of the utmost importance. They are made through face-to-face interactions and discussions. Even computer-generated plans are carefully discussed. Weak links in the supply chain are identified and limits placed on variations at each stage of the chain.
  • Practice PDCA—Plan, Do, Check, and Act—for the planning process itself is a method of taking corrective actions when a new problem is encountered. Changes in the planning process are made using the scientific principle.