| Oracle® Retail Demand Forecasting Implementation Guide Release 16.0 |
|
 Previous |
 Next |
Oracle Retail Demand Forecasting (RDF) is a statistical and promotional forecasting solution. It uses state-of-the-art modeling techniques to produce high quality forecasts with minimal human intervention. Forecasts produced by the RDF system enhance the retailer's supply-chain planning, allocation, and replenishment processes, enabling a profitable and customer-oriented approach to predicting and meeting product demand.
Today's progressive retail organizations know that store-level demand drives the supply chain. The ability to forecast consumer demand productively and accurately is vital to a retailer's success. The business requirements for consumer responsiveness mandate a forecasting system that more accurately forecasts at the point of sale, handles difficult demand patterns, forecasts promotions and other causal events, processes large numbers of forecasts, and minimizes the cost of human and computer resources.
Forecasting drives the business tasks of planning, replenishment, purchasing, and allocation. As forecasts become more accurate, businesses run more efficiently by buying the right inventory at the right time. This ultimately lowers inventory levels, improves safety stock requirements, improves customer service, and increases the company's profitability.
The competitive nature of business requires that retailers find ways to cut costs and improve profit margins. The accurate forecasting methodologies provided with RDF can provide tremendous benefits to businesses.
For a more detailed overview of the functionality within RDF, see the Oracle Retail Demand Forecasting User Guide.
This implementation guide addresses the following topics:
Chapter 1, "Introduction": Overview of the RDF business workflow and skills needed for implementation.
Chapter 2, "Implementation Considerations": Explanation of the factors to take into consideration before performing the implementation.
Chapter 1, "Introduction": Overview of integration and explanation of the RDF data flow and integration script.
Chapter 4, "Installation Considerations": Information for the setup that must be done prior to building the RDF domain and installing RDF
Chapter 1, "Introduction": Overview of the steps needed to setup and customize RDF.
Chapter 6, "Advanced RDF Configurations": Overview of the steps needed for advanced configurations in RDF.
Chapter 7, "Batch Processing": Information on the RDF batch forecast process.
Chapter 8, "Cross Promotion Effects Module (CPEM)": Information on the module for cross promotion effects.
Chapter 8, "Cross Promotion Effects Module (CPEM)": Information on the AutoSource utility.
Chapter 8, "Cross Promotion Effects Module (CPEM)": Information on the usage and configuration of Forecast Approval Alerts.
Chapter 12, "Internationalization": Translations provided for RDF.
Appendix A, "RPAS and RDF Integration with RMS": Details RMS to RDF transformation programs, RDF to RMS extract programs, Grade (RPAS) to RMS extract programs, and Curve (RPAS) to Allocation extract programs.
Appendix B, "Configuring the Cluster Procedure": Details how retailers can use Clustering to provide insight into how various parts of their operations can be grouped together.
Appendix C, "Configuring the Clone Procedure": Describes how Cloning can generate forecasts for new items and locations by copying, or cloning history, from other items and stores.
Appendix D, "AppFunctions": Describes how the AppFunctions library supports a number of functions and special expressions for RDF.
Appendix E, "Configuring the Preprocess Special Expression": Describes the Oracle Retail Preprocess module, Preprocessing, to correct past data points that represent unusual sales values that are not representative of a general demand pattern.
Appendix F, "Customizing Hooks for the RDF Generate Utility and Curvebatch": Details the hooks used in RDF and Curve for running customized computation at certain points of the batch process.
Appendix G, "Curve Configuration Process": Overview about Curve, an RPAS solution that is used to generate ratios from historical data at user-specified intersections
Appendix H, "Grade Configuration Process": Overview about Grade, a clustering tool that provides insight into how various parts of a retailer's operations can be grouped together.
Appendix I, "CPEM Calculations": Describes the special expressions that calculate cross promotional effects for Cross Promotional Effects Module (CPEM).
Appendix J, "RDF Script Names": Lists the revised script names common to RDF Cloud Service and RDF.
Appendix K, "Configuring the Forecast150 and CausalEstimate": Describes how to configure the Forecast150 procedure that generates the forecast and estimates the promotion effects.
Appendix L, "Configuring the Similarity Score Calculation": Describes the procedure to configure the Similarity Score Calculation.
Appendix M, "Appendix: Configuring Seasonal Profiles": Describes several options to create a seasonal profile for Curve.
Oracle Retail has designed a forecasting solution separate from replenishment, allocation or planning. In order to provide a single version of the truth, it is crucial to free up the user's time and supply the tools to focus on the analysis of forecast exceptions, historical data, and different modeling techniques. This empowers the user to make better decisions, thus improving overall accuracy and confidence in the results of the forecast demand.
Within the Oracle Retail Enterprise, Oracle Retail Merchandising System (RMS) supplies RDF with Point-of-Sale (POS) and hierarchy data that is used to create a forecast. Once the forecast is approved, it is exported to RMS in order to calculate a recommended order quantity. Forecasts can also be utilized (no export process required) in any Retail Predictive Application Server (RPAS) solution to support merchandise, financial, collaborative, and price planning processes.
Figure 1-1 shows the interaction between RDF, RPAS, and other applications.
Figure 1-2 shows the interaction between RDF Cloud Service, RPAS, and other applications.
One of the challenges in retail forecasting is the data volumes. The RDF business process focuses on automation, accuracy and lends itself to easy analysis. RDF focuses on automation by automatically selecting best forecast methods and parameters as well as by automatic approval of forecasts that don't meet any exception criteria. Also, it allows users to analyze and manually approve forecasts. Forecast scorecarding allows users to monitor forecast accuracy over time and re-tune settings if necessary.
Figure 1-3 illustrates the RDF business process workflow.
Following the initial setup, these parameters are not set on a scheduled basis, but are updated as needed.
Preprocessing and alert parameter setup
Sets forecast methods, parameters, and specifies source levels
Sets history start and end dates
For data pre-processing, RDF:
Corrects for lost sales due to stock-outs
Cleanses data for effects of promotions and short-term price changes (optional)
Manual data-scrubbing (fake history and user history overrides)
For new item / new store processing, RDF:
Calculates item similarity score based on item attributes
Generates recommendations for like item for new items
Clones sales history
For forecast generation, RDF:
Computes demand parameters (seasonality, level, trend)
Optimizes exponential smoothing parameters
Allows you to select best forecast method for item/location or use Automatic Exponential Smoothing (AutoES)
For exception processing and automatic approval, RDF:
Evaluates forecast for exceptions based on specific alert criteria
Automatically approves non-alerted forecasts
Allows you to review flagged exception forecasts
RDF provides the following features:
Pre-processing to correct for stock-outs and other data anomalies
Automatic recommendation of like item for new items
Generation of forecasts
Optimizes forecasting methods and exponential smoothing parameters
Selects best forecasting methods and parameters to overcome data sparsity and reliability issues
Generation of alerts and automatic approval of forecasts
Allows you to facilitate review of analysis and approval of forecasting
A typical RDF implementation team has technical and application/business consultants in addition to other team members.
The technical and application/business consultants need to have a high level understanding of other applications that RDF can integrate with, which include:
Advanced Inventory Planning (AIP)
Analytic Parameter Calculator for Regular Price Optimization (APC-RPO)
Analytic Parameter Calculator for Replenishment Optimization (APC-RO)
Replenishment Optimization (RO)
Retail Merchandising System (RMS)Retail Price Optimization (RPO)
In addition, both technical and application/business consultants need to have an understanding of RPAS, its calculation engine, and multi-dimensional database concepts.
|
Note: Staffing models and roles and responsibilities may vary from project to project, but following is a recommendation based on best practices. |
The technical consultant is usually responsible for the following key areas in addition to other activities:
Interface work
Batch scripting
RPAS/RDF domain partitioning
|
Note: The technical consultant should also be well versed in Unix, Shell scripting, and batch schedulers. |
The application/business consultant is responsible for:
Designing and configuring alerts
Configuring pre-processing rules
Any workflow/workbook customizations needed to meet retailers business process needs
|
Note: The application consultant should have a strong understanding of RPAS configuration rule language, RPAS Configuration Tools, RDF plug-in, and have experience configuring solutions on RPAS. |
