Demand Planning/ForecastingDemand Planning Systems, Demand Forecasting, Scheduling, Execution, Collaborative Planning, Forecasting and Replenishment, Sales & Operations Planning(S&OP) and improving quality control functionality.
IBM Partners with SAP on a Cognitive Demand Forecasting Solution for the Retail Industry
For retailers, improving their overall planning capabilities is a critical need in meeting consumer demand. Staying in-stock with accurate and timely forecasts while minimizing supply chain costs and excess inventories has never been more important. When it comes to planning, most retailers would like the ability to: Continuously improve merchandise planning Identify and gather relevant […]
For retailers, improving their overall planning capabilities is a critical need in meeting consumer demand. Staying in-stock with accurate and timely forecasts while minimizing supply chain costs and excess inventories has never been more important.
When it comes to planning, most retailers would like the ability to:
Continuously improve merchandise planning
Identify and gather relevant data from multiple sources
Comprehend the connection between consumer behavior, product demand, and external environmental variables
Identify and leverage unique correlations with demand influencing factors that would improve sales
To meet these requirements and deliver enhanced retail forecasting accuracy, IBM and SAP are teaming up to offer the IBM Cognitive Demand Forecasting (CDF) Solution, using machine learning and weather data to give retailers powerful new insights that will drive better business performance.
The forecasting engine for this new cognitive demand forecasting solution is provided by SAP’s Unified Demand Forecast (UDF). UDF is a component of the SAP Customer Activity Repository application, which collects client data for use in the retail industry. UDF provides demand modeling and demand forecasting services.
Demand modeling takes historical demand data provided as input, and then explains the historical sales and the impact that each demand influencing factor (DIF) had on consumer demand in the past. Examples of a DIF include price changes, promotions, tactics, seasonality, public holidays, or trend. Unified Demand Forecast then uses the results from demand modeling and is given inputs — such as planned promotions and prices — to predict the effects of similar DIF occurrences in the future. These inputs allow the system to derive future demand forecasts.
The IBM CDF Solution will provide hyper-local external data from IBM Weather Company to enhance the UDF modelling process. Weather plays a significant role in retail sales across many categories and can now be modeled up to seven months in the future. Using both artificial intelligence (AI) and weather data enables the CDF solution to continuously improve forecasts and the retail planning process.
The benefits include a precise and continuously improving forecast that enables:
Increased sales while avoiding unexpected product shortages and stock outs
Increased inventory level accuracy and allocation of merchandise
Enhanced planning, pricing, and promotion models based on product demands
The forecasting framework provides a flexible and configurable engine that allows retailers to determine the external demand influence factors to be recorded in the system and to be used at merchandise or sku level. For example, figure 1 shows the correlation between tomato soup and the weather information on a weekly basis.
The IBM CDF and the Unified Demand Forecast in SAP Customer Activity Repository are designed to automate the overall process to support merchants, planners, category managers, and the marketing department, equipping them with insights to improve their decision-making process. By leveraging improved forecasted results, retailers can now enhance their seasonal planning functions, including season merchandise and assortment planning, in-season store replenishments, and seasonal and weekly promotions planning.
With the prediction from the embedded Unified Demand Forecast in SAP Customer Activity Repository, the IBM CDF solution provides detailed analytical capabilities, which includes a detailed decomposition of the model and forecasted values. As result, marketing, planners, and merchants will have a completed picture that includes base historical sales, prices and promotions, channels, holidays, seasonality, and weather. Figure 2 is one example of the flexible decomposition analytical framework.
IBM CDF is based on innovative solutions from IBM and SAP, including SAP Leonardo, SAP Cloud Platform, SAP S/4HANA Retail for merchandise management, and SAP S/4HANA for fashion and vertical business, as well as The IBM Weather Company, to provide a unified and more accurate demand forecasting to retail for all retail planning functions based on machine learning. All this builds on omnichannel capabilities and deep industry insights from IBM Services and the SAP industry business unit