The Profitable Supply Chain: A Practitioner’s Guide (2015)
Chapter 5. Sales and Operations Planning
Sales and operations planning (S&OP) was formulated by Oliver Wight in the 1980s in order to align and synchronize activities within the executive team in a manufacturing company. As with other supply chain processes, the initial S&OP process blueprint has been adopted and modified to suit the individual requirements of each company. This chapter summarizes some common process and system approaches and describes the reporting and alerting tools that have proven to be useful across several industries.
The Importance of Sales and Operations Planning
S&OP is a cross-functional process that has gained in importance, largely owing to the realization that the performance of any department of a company is significantly influenced by events that can occur anywhere in the supply chain. For example, the sales department may find that revenue targets are not going to be met because a part is unavailable, or the finance department may have to deal with margin erosion because of excessive air shipments from the factory to the distribution centers. The realization of the importance of S&OP as a tool for improving performance can be seen in the following excerpt from an annual report of Lowe’s, a home improvement company.
In 2014, we will build on the momentum established in 2013 as we further optimize our business model. We will also continue to focus on three priorities to drive further top-line growth. First, we will use our enhanced Sales & Operations Planning process to improve seasonal planning by market. Second, we will improve our product and service offering for the Pro customer. Third, we will continue building customer experience design capabilities.
Through our Sales & Operations Planning process, we have addressed an opportunity to improve seasonal planning, including the cadence of product introductions, promotions, and staffing. While we have always planned and executed these seasons in store, previous planning was completed function-by-function and reconciled to minimize conflicts. Now the process starts earlier and is anchored on the customer mindset for the season. The process more thoroughly considers detailed input from all functions to determine resource allocation, and it enables Lowe’s to provide a consistent message and experience across all our selling channels.
—Lowe’s Companies, Inc., 2013 Annual Report
A similar account is provided by Avon, a consumer beauty products manufacturer:
We continue to implement a Sales and Operations Planning process that is intended to better align demand plans with our supply capabilities and provide us with earlier visibility to any potential supply issues.
Inventory levels increased during 2009, to $1,067.5 at December 31, 2009, from $1,007.9 at December 31, 2008, primarily reflecting the impact of foreign exchange and business growth offset by operational improvements. New inventory lifecycle management processes leveraged with initiatives such as PLS, SSI, ERP implementation, and the Sales and Operations Planning process are expected to improve inventory levels in the long term. Inventory days are up seven days in 2009 as compared to 2008, due to the impact of foreign exchange. We expect our initiatives to help us deliver operational improvements of three to five inventory day reductions per year for the next two to three years.
—Avon Products Inc., 2009 Annual Report
The growing importance of S&OP may be partly attributed to the success of supply chain management and its focus on connecting demand and supply across the entire network. Some of the questions addressed by an effective S&OP process include:
· Are there any issues in the supply chain that can affect the company’s performance related to revenues, customer service, lead times, and margins?
· What options are available to resolve issues? Who needs to be involved in the decision-making process? What are the cost implications of these actions?
The remaining sections of the chapter address these questions by providing you with details related to workflow, as well as worksheets that you can use to analyze issues and provide resolution options.
The S&OP Workflow
S&OP consists of a set of activities to organize demand and supply information, identify exceptions (issues) requiring a discussion with other divisions in the company, present these issues and resolution options to management, and execute the decisions that are taken (see Figure 5-1). Even though S&OP is cross-functional, it is important to assign an owner to this process in order to ensure that all issues are being adequately addressed and that the desired goals are being met. It is natural for the owner to be from the supply chain department, if a company is organized accordingly. Otherwise, an owner needs to be picked from one of the other departments, usually manufacturing or procurement.
Figure 5-1. The sales and operations planning (S&OP) process
The most common practice is to perform S&OP on a monthly basis, with weekly reviews involving the parties that have been tasked to take action. If any of these actions cannot be completed or if cost overruns are identified, then additional meetings with executives can be scheduled on an as-needed basis.
S&OP Procedures and Templates
The critical items that need to be reviewed during S&OP include:
· Customer order exceptions. Whether any customer orders are in danger of being delayed.
· Demand forecast and revenue exceptions. Whether forecast volume or revenue are in danger of not being met.
· Inventory shortfall exceptions. When supplies are insufficient and demand cannot be met.
· Margin exceptions. Whether any cost overruns are anticipated.
· Cash exceptions. Whether inventory is projected to be in excess of targets, leading to an increase in invested capital.
Each of these exceptions can be the result of very different causes, including insufficient raw material inventory, delays in transportation, or insufficient capacity. Another complicating factor is that the physical network can influence the number of points at which issues can occur, such as whether the company ships directly from factories to customer, or if there are one or more distribution centers. Therefore, methods for reviewing and analyzing each of these exceptions depend on a combination of several factors, a few examples of which are provided in the following section.
Customer Order Exceptions
Shipment delays can be analyzed by comparing ship dates with projected inventories, which, in turn, are computed based on the on-hand and planned supplies. An example of a spreadsheet view that provides this comparison is shown in Figure 5-2. In the diagram, the first view is a summary of the exceptions ordered by the severity of the issue. This summary view provides guidance regarding which issues need to be addressed first, which is helpful when the number of lines are numerous.
Figure 5-2. A sample template for analyzing customer order exceptions
This summary view lists important information related to the customer account, the site from which the order is to be shipped, the planned shipment date for the order, order summary (units and financial), the number of line items in the order, shortage information (the number of items not shipping on time, the number of units not shipping on time, and the financial value of the short units), and the year-to-date (YTD) shipping performance of the customer. This information can be used to gauge the severity of the issue and any potential risk associated with a customer. Along with the columns shown in Figure 5-2, additional columns listing more information such as the names of the short items and reasons are provided to aid the troubleshooting procedure.
In this example, goods ship out of a distribution center, which, in turn, is replenished via truckloads from the factory. The detailed view for decision-making requires a daily view of orders, shipments, and inventories, and a sample table that provides the necessary information is shown in the daily inventory view in Figure 5-2. Details include:
· Firm orders: customer orders with a firm ship date (latest). Any delays to this ship date may be considered a delay.
· On-hand inventory: the inventory at the time of review.
· Receipts: the expected receipts for any particular day.
· Projected Inventory EOD: the end-of-day projected inventory, calculated as the beginning inventory plus receipts minus shipments.
· Manufacturing (MFG) block: repeated fields are omitted in the description.
· Shipments: the demand placed on the manufacturer (i.e., shipments to the distribution centers).
· Production: the planned production for the particular item.
Specific shipment and receipt information are provided in the bottom block, with specifics regarding dates, carriers, status, delays, and reasons. With this information, it is possible to analyze the situation and options. The negative projected inventory in the first block, for example, indicates that the in-bound shipment is arriving late and there is insufficient inventory to fulfill the order completely. Since the shortfall can be fulfilled by the receipt expected in two days, the order will be fulfilled with a two-day delay. Similarly, the 25-unit order on Feb. 10 can be fulfilled only by Feb. 14, resulting in a four-day delay. Since there is sufficient on-hand inventory at the plant on Feb. 8, there is an option available to ship the 25 units using expedited transportation (overnight delivery) and have the goods available in time to be processed and shipped. The additional cost for expediting can be discussed during the S&OP meeting, and a decision taken based on the margin impact.
Revenue exceptions can be analyzed by comparing the demand forecast against inventories and planned production. This analysis can be performed at a monthly level, with quarterly summaries to compare against sales budgets and revenue targets. This exception can be reviewed at the product-line level at which targets are established, which helps reduce the volume of data that needs to be analyzed.
There are several ways in which sales vs. forecast can be analyzed, depending on the sales model and demand patterns in a month or quarter. For example, the analysis can be made assuming that sales come in evenly over the quarter, such that the duration of the quarter that has elapsed can be compared to forecast achievement to flag an exception, as shown in Figure 5-3. The important fields are explained as follow:
· % of Qtr elapsed. Given the current date, the duration of the quarter that has elapsed can be calculated as the number of days in the quarter elapsed divided by the total number of days in the quarter.
· Target forecast achievement. This is the target level of sales (as a % of forecast) that need to have been realized. The 41% value in the example is identical to the duration of the quarter that has elapsed, since a linear sales profile has been assumed. If a company is sales-heavy toward the end of the quarter, this value can be modified downward to reflect such a pattern. However, even in such cases, it is recommended that a linear pattern be used in order to ensure that sufficient inventory will be available in case orders come in earlier than expected.
· Required sales rate. This is the revenue that remains to be realized in the quarter, divided by the number of remaining days. This rate can be compared to the sales rate achieved during the prior quarter as well as the same quarter in the previous year; such a comparison will help estimate if the required sales rate seems achievable.
· Product line-specific details, along with a summary of reasons for shortfalls. These reasons can help guide the S&OP team toward a resolution procedure. For example, if the reason for the shortfall is insufficient supplies, then the supplier situation needs to be analyzed. However, if the reason is due to light sales, then promotions can be evaluated.
Figure 5-3. A sample template for analyzing revenue exceptions
Inventory Shortfall Exceptions
When supplies and on-hand inventory are not sufficient for meeting projected demand, the first resolution step to be considered is expediting supplies. When it is not possible to make supplies available for a certain period of time, then the company has to consider allocation options. This step is often performed in an unstructured, ad hoc manner, often assigning scarce material to the first sales representative requesting material, or to the first sales order on the list. However, the proper use of scarce resources is extremely important, and it behooves a company to implement a structured procedure for performing allocations. A few of the considerations and methods for performing allocations are as follow:
· Contractual obligations may specify a service level or quantity that needs to be fulfilled, with a violation resulting in a monetary penalty. Therefore, it is important that when a constraint situation is identified, the quantity required to fulfill contractual obligations be calculated and reserved.
· Strategic considerations can include important customers, markets, geographies, and products. Here, importance is a subjective term, and its definition can vary significantly across companies. For example, customers can be classified as strategic due to high volumes or growth potential. Markets and geographies can be considered strategic if they are new or the company is engaged in intense competitive activity. Similarly, products can be considered strategic based on margins and growth potential.
Two broad allocation procedures need to be considered: The first is order allocation, which is the method by which supplies are allocated to customer orders. If there is insufficient inventory to satisfy all the orders, allocations can be performed according to margins, first-come-first-serve(FCFS), or fair share (where the available units are distributed across the orders in proportion to ordered quantity, with modifications for batch sizes).
Forecast allocation is the method by which supplies are allocated to the forecast provided by different sales accounts or regions. Since demand may come in very differently from the forecast, it is possible for poor business decisions to be made during forecast allocation. For example, scarce supplies may be allocated to a particular region based on the forecast provided, only to find that sales did not materialize, while sales in another region have outstripped supplies. For this reason, additional considerations and allocation methods are required.
Forecast allocation can be performed according to margins provided by each region or account, or based on a fair share approach that allocates supplies to each of the regions according to the provided forecast. Fair share is preferred by many companies since it maintains a certain market presence and customer service level in all the regions; on the contrary, allocating based on margins may result in one or more regions receiving no supplies at all.
However, a drawback with fair share is that it is susceptible to manipulation by the sales organization. If the regional sales organization deliberately increases the forecast in order to increase allocated supply, fair share can result in unused inventory, which is expensive when supply is constrained. This situation can arise even with customers, as explained by Maxtor, a manufacturer of hard disks for computers:
Backlog: The Company generally sells standard products according to standard purchase order terms. Delivery dates are specified by purchase orders. Such orders may be subject to change or cancellation by the customer without significant penalties. The quantity actually purchased and shipment schedules are frequently revised to reflect changes in the customer’s needs. At times, when price competition is intense and price moves are frequent, the Company believes most customers may place purchase orders below their projected needs, delay placing orders, or even cancel purchase orders with the expectation that future price reductions may occur. Conversely, at times when industry-wide production is believed to be insufficient to meet demand, the Company believes that certain customers may place purchase orders beyond their projected needs in order to maintain a greater portion of product allocation.
—Maxtor Corp., 1999 Annual Report
Although this is a 1999 report, the same business practices and challenges continue to be faced by companies even today.
Clearly, a simple product allocation method can result in the positioning of inventory at the wrong places, unless it is possible to design the allocation method to take this behavior into consideration. One such method is the performance adjusted fair share, which tracks the supply utilization ratio (i.e., the effectiveness of each region in utilizing allocated supplies). The ratio of actual sales to the available supply is used as a measure of performance, as illustrated in Figure 5-4.
Figure 5-4. A sample template for analyzing the effectiveness of utilizing allocated supplies
The higher the value, the more effective the sales organization is at utilizing the provided supplies, while a low value indicates that the region is inflating forecasts artificially. With this effectiveness measure, the allocation for a particular period can be calculated as follows:
Finally, if lead time permits, scarce supplies can be retained as available to promise (ATP) and can be assigned to individual orders as they are placed. This method is popular in companies with numerous sales individuals, as well as Internet retailers that take orders directly from consumers and ship from a central distribution facility. The most common method for assigning ATP to individual orders is first-come-first-served (FCFS). In situations involving high transaction volumes, use of ATP will require the development of a software application to manage the assignment of inventory and calculation of remaining supply. The use of ATP is common in online retailing, where available inventory is allocated to a customer’s shopping cart for a specific period of time (say, 1 hour). If the customer has not placed a firm order within this time, the allocation is released and made available-to-promise for other shoppers.
Depending on the type of fulfillment, it may be necessary to utilize one or more of the above allocation methods. Therefore, it is important to follow a well-defined procedure for allocation, as shown in Figure 5-5. The steps specified are to identify the supply constraint, followed by steps to reserve inventory to meet obligations, and allocating any remaining inventory to forecasts. Such a rigorous procedure can ensure that scarce resources are being utilized to meet company objectives in the best possible way.
Figure 5-5. A process for allocating scarce supplies
Excess Inventory Exceptions
Like the low inventory situation examined in the previous two cases, excess inventory indicates a mismatch between demand and supply, but with supply in excess of demand. The challenges in reducing supply to match lighter sales include:
· Supplies may already be in-transit and therefore cannot be reduced.
· Production may have been initiated and therefore, the inventory may be close to completion (work-in-process).
· Purchase orders may have already been placed with suppliers, and the company may incur a penalty for canceling these orders.
· Reducing planned production may result in lower output, resulting in higher costs due to fixed costs being absorbed by fewer units.
Therefore, excess inventory situations have to be analyzed in detail to understand the cost implications of each of the resolution options. The forecast waterfall report, described in Chapter 7, is a useful visual aid for understanding if the excess inventory is due to over-forecasts.
Because the S&OP process is very data-driven, the ability to present relevant information in a succinct manner is very important. However, most software applications continue to rely on simply presenting large volumes of data to the analyst, and the onus of understanding issues and causes still remains with the human. This section provides a few ideas and guidelines regarding the presentation of information in a manner that is suitable for analysis and decision-making. My intention is that you build on these ideas to suit the needs of your review group and company.
Some of the concepts presented in this section come from the field of data graphics, which studies ways of displaying numbers to aid reasoning.1 The most fundamental graphic that substitutes for a table of numbers is the line graph. But even this simple graph needs to be created carefully.Figure 5-6 shows how poor design and formatting can impede rapid analysis. In the example, fill rates are plotted along with the diagnostic measures, forecast accuracy, and days of inventory. Clearly, the graph is not easy to read and it is hard to gain any insight into the issues and causes. Several problems in the way the graph is constructed contribute to this drawback: The use of dual axes (for percentage and days of inventory) is often confusing. Additionally, overlaying multiple series in the same graph increases the time required to analyze each individual series and can mask trends.
Figure 5-6. A graphical design that impedes rapid analysis
The same information can be presented in a different manner, with the different series being plotted in separate graphs, as shown in Figure 5-7. While the space occupied has increased, it is easy for the analyst to track trends for each series and to perceive relationships between variables. In this situation, the second graph does a superior job of quickly indicating that the dip in service levels in June coincides with a reduction in inventory as opposed to forecast accuracy.
Figure 5-7. An improved design for presenting information
Indeed, several other designs can be equally or more capable of providing the necessary insight. Unfortunately, effort is required to implement these designs using standard software packages and spreadsheets, which, by default, tend to produce graphs like the one shown in Figure 5-6.
Another important requirement is to capture supporting information necessary to provide a complete picture, to explain why there was deterioration in the observed metrics, and to provide supporting information and comments from the various participants. A visual aid that combines several of these design elements is the metric information chart, which has the following characteristics:
· A combination of metrics that are required to provide clarity regarding a situation. The previous example combined fill rates with inventory levels and forecast accuracy.
· Acceptable ranges for the metric, highlighting upper- and lower-control limits. The ranges help focus attention on significant changes in performance.
· Supporting data, including specific product and item details, to provide context for analysis and decision-making, and specifics so that the analyst has an initial regarding the product lines or products that need to be further examined.
· A combination of quantitative and qualitative reasons for issues. The qualitative factors are often contributed by different organizations in the company. This important addition to the information collected helps increase process memory–information regarding why quantitative methods were overridden, what additional causal factors needed to be considered, and important business constraints can all be captured.
· A list of actions taken, issues faced, results, and benefits obtained.
The example provided in Figure 5-7 is now modified to include these additional elements, and is illustrated in Figure 5-8. The graph has been augmented to include a lower bound for fill rates to highlight dips that require attention, and annotated to highlight issues, comments, and actions for future reference and tracking actions.
Figure 5-8. An example of a metric information chart
Other examples of visual aids include the waterfall chart for forecast accuracy, a histogram for tracking forecast bias, segmentation charts for ranking customers and suppliers, and maps for tracking network performance. However, the analyst should be careful to use the right balance of graphs and other traditional methods, such as tables. There are several situations in which a simple table is more effective at communicating the necessary information, as shown in Figure 5-9. In this example, cost information is summarized using a waterfall chart and a table. While the waterfall chart is visually pleasing, it is occupies more space than the table and does not convey variance information. Therefore, the table is a more effective analytical tool in this case—and in the general case in which just a few numbers and a lot of text need to be displayed.
Figure 5-9. Effectiveness of a table for displaying limited data and supporting information
The approach for determining the best visual format and right balance between presenting too little or too much information is an iterative process. The analyst needs to create an initial format and present the data to the meeting participants. The feedback and additional questions from the audience can be used to improve the presentation. It may take as many as three or four iterations before participants are satisfied with the format. Even after this, additional changes may be needed as new exceptions or new information continues to be included in the analysis.
Sales and operations planning is an important process that involves several departments of a company—manufacturing, procurement, distribution, sales, marketing, and finance. Representatives from these departments are responsible for communicating changes to demand and supply, providing causal insight, and recommending actions that best alleviate the situations. There are several challenges faced while implementing this process, such as obtaining the support from the different departments and ensuring that the stakeholders devote sufficient time to prepare for these meetings. However, the biggest challenge is being able to assimilate and present the necessary information in an easy-to-understand format so that decisions can be quickly taken. Once participants begin to see performance improvements, it is a lot easier to gain the internal support to address the other challenges.
1Tufte, E. R., The Visual Display of Quantitative Information, 2nd Edition, Graphics Press, 2001.