A Complete Guide to Forecast Improvement in SAP Integrated Business Planning
“Improve Forecast accuracy to gain sales, reduction of inventory and as consequences increase customer satisfaction.“
The blog covers all the aspects of forecast implementation. It aims in helping business owners who are looking to fulfil their future demand profitably with real-time supply chain management. Moreover, business owners can learn how to optimize your supply chain and create value for their company.Read this guide and enable your company to save up to 100 Million Euro (scenario of 500 Million sales). An improvement in forecast accuracy gives you significantly more leeway:
- Reduction of stock levels and lower stockholding costs
- Increasing your service level and customer satisfaction
- Production of the right products at the right time
At SCPLAN GmbH we apply a multi-level forecast for this purpose, which improves the forecast quality by 5 – 20 % on average. The forecast results include different levels for the best possible forecast outcome.
With Digital Transformation changing at a rapid pace, it is difficult for top businesses to manage and adapt to the change swiftly. After managing high market volatility, shorter product life cycle, and variation in demand, the delivery lead time still remains competitive. Further along, there are various other daily work challenges like inefficient operations, inadequate forecast accuracy, excessive inventory, and unwanted low service level that affect sustainability goals. All in all, the demand forecasting has become an unmanageable task for business owners to perform with satisfactory results.
This is where Forecast improvements comes into the picture. It has helped top industry leaders in establishing long-term partnerships with their most strategic customers. With this, they are easily managing their growth and sustainability goals with a strong focus on the supply chain.
Thus, Forecast improvements in SAP IBP propose new opportunities to adapt to the new normal context.
Let’s dig deep into what forecasting is and how it can benefit your Supply Chain:
The forecast process is a key part of Supply Chain. It relies on statistical algorithms to interpret historical data and extrapolate it into the future. To project demand based on the market, it is recommended to use an as input for forecast the requested delivery date (RDD) and requested delivery Quantity (RDQ).
Keeping the right amount of product in stock at the right time is critical to businesses. Having too little means running out at inopportune times, causing customers to buy elsewhere. Also, having too much means paying unnecessarily high costs for storage and inventory management.
When you implement forecasting in your supply chain business, you can be ensured that you have enough supply on hand to satisfy demand. Here are some of the key benefits of Forecast improvements that you can consider before making any decision:
- Increased Service Level: Various customer service metrics like on-time-in-full(OTIF), on-time delivery(OTD), case fill rate are improved with forecast implementation. Hence, the right-sizing and positioning of inventory results in enhanced distribution planning and optimized inventory levels.
- Optimization of Inventory Levels: A proper Forecast gives requisite data for managing the desired raw material, WIP as well as finished goods inventory levels. This decreases the chances of the Bullwhip effect across the Supply Chain. Further leading to the optimization of inventory levels and reduction in stock-out or over-stocking situations.
- Improved Supplier Relations and Purchasing Terms: Forecasting pushes the raw material planning process timely. It further facilities the Purchasing Managers to release appropriate purchase plans to suppliers. The transparency and visibility of raw material demand thus enhance supplier relations and enables Purchasing Managers to negotiate beneficial terms for their companies.
- Better Capacity Utilization and Allocation of Resources: With improved forecast, the production can be scheduled efficiently and effectively. The transparency of the current inventory levels, expected customer orders, and raw material availability contribute to it. All in all, this leads to enhanced capacity utilization and accurate allocation of manufacturing supplies.
- Improved Distribution Planning and Logistics: Not only small businesses but businesses dealing with multiple SKUs and wide distribution networks can balance inventory across the network easily and can negotiate with the transporters on favorable terms.
- Facilitates Performance Management: With forecast implementation, organizational efficiency, effectiveness, and improvement initiatives can be designed for key areas of the company. Management can set KPIs and targets for several functions like Manufacturing, Sales, Purchase, Finance, Logistics, etc. based on the medium to long-range plans derived from the Forecasting process.
- Better Product Lifecycle Management: Forecasts render a better visibility of both the new product launches and old product discontinuations. This induces well-planned raw material, manufacturing, and inventory planning in order to promote new product launches and most importantly, decreasing the risk of obsolescence of abandoned products.
Key aspects of forecasting are segmentation and lifecycle planning. At any time the planner needs to know about the priority based on revenue and the volatility of planning data. Within this whitepaper we will explode the segmentation and lifecycle planning.
Supply Chain planners are overloaded with a lot of information, which makes decision making a challenge. Segmentation enables grouping of planning objects based on data and thereby increase user efficiency and planning accuracy. It is used in different area like analytics and custom alerts.
SAP IBP provide Segmentation based on revenue (alternative volume) and volatility. It checks the planning data and determines from the highest revenue to the lowest and from very stable to high volatile sales. Those informations are grouped and called segments. Purpose of segmentation is to priorities all products, which has a high value for the company. Furthermore the volatity tells a planner, which products are easy to forecast and where to focus most. From planners perspective to will helpl to identify, which SKUs are most important for the company. Based on this information it is easy to priorities on most important task on a daily level.
Overview of ABC, XYZ Segmentation:
Based practical experience a segmentation is required on product and product location level. Furthermore from sales perspective to identify your priority of customers, consider to segment as well based on customer and country. In addition a separation of division is in some cases is required, when Business sees big variation in weight of the division.
ABC analysis is performed to classify planning objects according to their usage value, or several objects. During ABC analysis, the system allows each object one of the following indicators:
- A – The most value, or given number of objects that produce the greatest value
- B – Less value, or given number of objects that produce less value
- C – The least value, or given number of objects that produce the least value
XYZ analysis is performed to classify planning objects according to the variance in a particular coefficient. During XYZ analysis, the system assigns each object to one of the following indicators:
- X – A very little variation
- Y – Some variation
- Z – The most variation
Phase-In / Phase-Out is part of lifecycle Planning (PLC). It covers following phases launch, growth, maturity and discontinuation.
Since every stock Keeping Unit is an Actual SKU, you want to label the “updated/new” SKU as the SKU you want to forecast whereas you want to label the future SKU as the “Planned SKU.” Most of the forecasting software these days allows you to map the old SKUs to the new SKU and call it mapping Planned to Actual. This way, you can label the Planned SKU as Phase-In, and the Actual SKUs mapped to it as Phase-Out.
All in all, if you identify old SKUs to sell first and communicate these inventories weekly, you indicate selling the old SKUs first and avoid working capital costs to write off inventory. You can communicate this list and hence show the progress, that you have saved.
Phase-out of products is a necessary step to stop generating forecast and therefore ensure no more capital lost for inventory and disposal.
Life cycle Planning
New products starts with a ramp-up, which is called a launch in the market. From the launch a rapid growth is demanded by the market. When the product is on the market for longer period the sales is stable. However after a while a product will be discontinued due to innovation and customer requirements. Therefore a decline of demand is see and companies needs to react quick enough.
With delayed shipment, inconsistent suppliers, and inefficient plants, companies across various countries are facing complex challenges. With forecast improvement, companies can overcome the following mentioned challenges:
Seasonal forecasting plays a major role in seasonality. It includes managing the products that have higher demand during a specific time of year. This can help businesses in managing for these peak season and produce sufficient materials to build the final product and have enough inventory to fulfil the demand given at that time of the year.
However, there are some uncertainties that can still happen because of unforeseen events that can occur during a specific year. There is no accurate way to predict these types of events but is a financial risk that a supply chain manager may make. The thing is the majority of the time, these types of events are rare so it does make more sense for managers to go ahead preparing for the season rather than having not enough products to sell during the peak demand. Having an extra inventory is not always a bad approach compared to the option of not having the supply to meeting the demand and missing that window of opportunity to optimizing profits.
By the above-explained point it is clear that forecasting takes into account many changes, such as demand seasonality, by tracing patterns in historical data. Coming to the other challenge – Outliers. They are the points in historical data that diverge from other data points naturally. They can be either be large or small. They are not considered a significant part of the overall demand pattern because they lie far outside of the expected data range. It is one of the main challenges of forecast improvements.
Since, Outliers can be irregular, but they also can occur due to various planned events, such as sales promotions and unplanned ones, including competitor promotions and natural disasters.
- Intermittent Demand:
It has been observed that as many as 50% of products and services have demand patterns with “lots of zeroes”, which in return creates different challenges for demand estimation. It leads to the failure to handle “lots of zeroes” correctly can cripple the effectiveness of an operational process in a supply chain business.
To avoid this, it can be vital to divide the products with “lots of zeroes” into two groups:
- Structural zeroes– These zeroes have a pattern that relates to the structure of the supply chain or data collection methods.
- Intermittent / Sparse / Lumpy– It has lots of zeros spread randomly across time.
All in all, demand with lots of zeroes requires special attention and expertise. The idea here is to consider the estimation process as a risk trade-off. Firms that do this with a proper strategy can see a large improvement in performance.
- Human Judgment:
Statistical forecast is generated automatically by the system. It is the ground for further review, adjustment and alignment. At the end of Demand Planning process, the consensus Demand is published. It includes a lot of human interaction and can lead to wrong conclusion. Therefore, a measurement how good is the system versus how good is the forecast result of a human is supporting to continue usage of system intelligence.
The business might be surprised when we say, that best forecast result is not generated by spending a lot of time of finetuning forecast parameters, method, considering weather condition etc. Each mentioned point is important, nevertheless the focus should be on forecasting on multi level. This ensures that the overall numbers on aggregated level, which considers the historical data on e.g. product hierarchy and provides a very good outcome to project the demand for the same product hierarchy. Since each project is individual, it is best practise to determine for each client individually on which level to get the best aggregated forecast.
Furthermore, for detail level for instance on product and customer level a forecast is generated. This forecast will be used to disaggregate from product family to product and customer. By this approach we successfully implemented an improvement between 5 to 20 percent. Our clients benefit from it EVERY YEAR.
Forecast improvements is impacted by several factors. Each Business has its own view of important inputs. Therefore most common aspect to improve forecast will be explained below.
- Segmentation: classify your product portfolio with segmentation in ABC and XYZ. This supports to identify, how high is the forecast accuracy and forecast error on each matrix. Most clients use the combination of ABC and XYZ for instance A products in combination with stable sales called X will use it AX. Furthermore they want to know forecast accuracy on matrix level.
- Lifecycle Planning: Improvements of forecast goes hand in hand with product lifecycle planning. Since the launch of new products needs to be organized as well as for discontinuation of products.
SAP IBP allows first to assign the predecessor with the successor:
At the next step forecast dates needs to be maintained.
- Forecasting Multilevel: Multi-level forecasting is the capability to produce sales forecasts on the detail level, of both the products and customers via:
- Detail forecasting – at sku + customer level, which will be used for disaggregation
- Top-down forecasting – at the Total level e.g. Product Family/hierarchy and spread down to product and customer level.
- Result forecast is available after Top-down forecast disaggregated based on detail level
- Aggregated: With aggregate forecasting, a company’s capacity requirements i.e. the amount of product that is required to produce along with the strategies for producing it for a period of 2 to 12 months in the future are considered. A business can set or reset strategy in this period and adjust the activities that affect resource allocation, costs, capacity utilization, labor requirements, and customer relations.
- Disaggregation: It is a method of breaking an aggregate plan into greater detail. For an example a product group consists of two products, forecast on aggregated level will be disaggregated on lowest level, means on each product.
- Final Forecast Result is saved after Top-down to the lowest defined level and will be used for further stages like forecast adjustment.
Forecasting Multilevel overview:
SAP IBP is providing many templates for analytics and dashboards, which can be used as a starting point. Especially for forecast improvements KPI reports are easy to establish for example by using segmentation and forecast accuracy or Forecast error. This report can be compared every demand cycle and planner can learn from experience in the past.
Dashboard example for Forecast accuracy:
Besides the matrix accuracy view, planner need as well the consider the patterns in combination with the segmentation. Important information for instance about how many seasonal pattern where recognized for A products can be displayed, see example below:
Is your Business ready for forecast improvements? When do you want to establish savings YEAR after YEAR?
We are ready to support your business to create value for your supply chain.
For more information, please reach out SCPLAN GmbH:
Get in touch…Contact