This chapter covers the following topics:
The Oracle Demantra applications suite contains these products:
Oracle Demantra Demand Management
Oracle Demantra Advanced Forecasting and Demand Modeling
Oracle Demantra Real-time Sales and Operations Planning
Oracle Demantra Predictive Trade Planning
Oracle Demantra Trade Promotion Optimization
Oracle Demantra Deductions and Settlement Management
For detailed information and business examples about the features and functions, see the Oracle Demantra documentation for each product.
Oracle Demantra Demand Management enables an organization to accurately forecast and plan for demand. It uses state-of-the-art Bayesian Markov forecasting, as well as pre-seeded hierarchies, worksheets, and workflows to establish a baseline forecast based on a combination of quantitative or qualitative data. Demand Management offers unlimited hierarchies for multi-dimensional analysis with dynamic reporting capabilities.
Reporting And Analysis Features
Predefined Demand Management worksheets:
Compare and analyze sales history and forecast demand
Aggregate data in various ways to determine forecast accuracy of previous sales cycles
Manage and approve demand forecasts
Standard and predefined:
Hierarchies (levels): Used in worksheets to determine how data is aggregated and organized. Enables users to analyze data at various levels, such as by customer, product, location, and demand class.
Series: Predefined set of data displayed in a report, table, or graph, at any aggregation level. Included are primary key indicators, forecasts, and accuracy metrics.
Slice-and-dice data capabilities:
Any dimension and level of granularity
Customize data views by aggregating and organizing in hierarchies, currencies, and units of measure
Unlimited hierarchies, levels, and attributes
Workflows, role-based processes that automate specific business flows and tasks:
Managed forecast approval process
Integration with Oracle J.D. Edwards Enterprise One (E1) and the Oracle E-Business Suite (EBS)
Other Features
Advanced statistical forecasting, including up to 20 causal factors.
Collaboration platform, includes alerts and exceptions.
Evaluate demand from multiple data sources, for example, shipment and order history and sales and marketing data.
Nine forecasting models within the Bayesian framework.
Offline capabilities.
Oracle Demantra Advanced Forecasting and Demand Modeling is an add-on to Oracle Demantra Demand Management. It has robust and powerful modeling tools that you can use to address complex analytical questions and provide a complete demand forecast:
Advanced Statistical Analysis (base and lift decomposition): Decompose the forecast into baseline and lift. See forecast for promoted periods separated into baseline volume and volume associated with the promotion.
Advanced Lift Decomposition: See what the effects of running a promotion on a product and location have on other products and locations. This type of analysis breaks down the impact of an event into:
Baseline (unpromoted volumes)
Direct lift
Pre-effect (event anticipation)
Post-effect (event driven pantry loading)
Halo and cannibalization for four different switching types
Item
Brand
Store
Channel
Cross-correlative analysis
Unlimited causal factors: With AFDM you can define an unlimited number of causal factors.
Localized configuration of the Analytical Engine (nodal tuning): Fine tune analytical engine selected models and parameters for a sub-set of items and locations. See Performing Advanced Analytics in the Oracle Demantra Demand Management User Guide.
Shape modeling: A statistical estimation of product launch rates (shape) based on prior learning. For details, refer to the Oracle Demantra Implementation Guide.
Promotional and sales calendarization: Normalizing to standard calendar periods.
Advanced reporting: Gantt charts for deals, trade, promotions, and price breaks and other events.
Advanced logistic forecasting and ridge regression-type forecasting models (normally used with large numbers of causal factors). These include:
Integrated Auto and Linear Regression (ARIX)
Auto Regressive Logistic (ARLOGISTIC)
Dual Multiplicative Linear Regression (DMULT)
Multiplicative Monte Carlo Regression for Intermittent (ICMREGR)
Logistic (LOGISTIC)
Modified Ridge Regression (MRIDGE)
The general procedure for configuring AFDM for a promotion is as follows:
Review the following sections in the Oracle Demantra Analytical Engine Guide:
Introduction to the Analytical Engine
Configuring the Analytical Engine
Review the following sections in the Oracle Demantra Implementation Guide:
Configuring Promotion Effectiveness
Configuring Promotions and Promotional Causal Factors
Promotional Lift Coefficients
Forecast Decomposition (PE Mode Only)
Define and populate your promotional causal factors (see Configuring Promotional Causal Factors in the Oracle Demantra Implementation Guide)
Enable the AFDM module for each relevant user. The Demantra system administrator does this in Business Modeler by navigating to Security > Create/Modify User, and then selecting AFDM in the User Modules section.
Enabling forecast decomposition
Set the RUNMODE parameter to 1 (in init_params_0), and then run the analytical engine. The forecast will include the baseline forecast plus the defined uplift.
Add the relevant promotional series to your worksheets. For example, search the list of available series names for those containing “lift,” “Canbl”, "cannibalization," and "uplift" (uplift is the most important).
Ensure security allows open analytics to be viewed by users (see step 3, above).
Use the Open Analytics option to configure models and engine parameters, per specific items and locations.
For details, see:
Advanced Analytics (Nodal Tuning)
Performing Advanced Analytics (Nodal Tuning), in the Oracle Demantra Demand Management User Guide
Enabling shape modeling
Create causal factor of type 'Activity'
Add shape-specific series to worksheet and specify timing of shape.
For details, see About Activity Shape Modeling.
Enabling advanced analytical models
Activate advanced models globally using Business Modeler (refer to the list of models above, in AFDM Features).
For details, see Enabling Engine Models Globally.
Activate advanced models locally using Open Analytics.
For details, see:
Advanced Analytics (Nodal Tuning)
Performing Advanced Analytics (Nodal Tuning), in the Oracle Demantra Demand Management User Guide
This application provides a collaborative forum in which an organization's sales, manufacturing, and financial groups work together to reach demand consensus and create a unified business plan.
Predefined Templates Feature
Predefined templates are predefined, function-specific worksheets enable those involved in the sales and operations planning process to collect sales, marketing, manufacturing, and financial data. This data then drives the following collaborative business processes:
Demand Plan Review: Stakeholders--typically the marketing and sales departments--review the company's forecast of future demand and use predefined reports to create a Consensus Demand Forecast. This forecast is an input to the Supply Plan Review.
Supply Plan Review: Stakeholders review the Consensus Demand Forecast and then use appropriate reports to create a plan for meeting the expected demand. This information is used to generate a Consensus Demand Plan. This plan is an input to the Executive Review.
Executive Review: Senior management uses relevant reports to compare planning data with the company's budget and reviews key performance indicators. The result is a set of consensus demand and supply plans that drive enterprise resource planning (ERP).
Other Features
Predefined workflows:
Task notification and approval process for Demand Review and Supply Review
Preconfigured series and calculations
Oracle Demantra Anywhere: Pure HTML view of data
Dynamic Open Link to Excel to create reports
Enhanced functions for modeling supply side series, calculations, and parameters
Fpos and Fsum functions to:
Model supply-side series, calculations, and parameters
Launch workflow-based automation and management approval process
Oracle Demantra Predictive Trade Planning has a fully functioning trade promotion management capability for trade funds, promotion entry, execution, tracking, and analysis. It uses advanced data mining technology so your organization cananalyze sales history, efficiently manage trade funds allocation, and understand how promotional strategies shape demand.
Features
Post-event analysis: Analyze overall promotion effectiveness
Predict the impact and profitability of future promotions
Predict and monitor trade fund spending
Manage all funds, sales, and account planning:
Supports multiple fund types and budgets
Enter funds and rates by territory
Allocate funds by territory, retailer, brand, or promotion group
Create profitability reports to track costs and benefits of events.
Easily view forecast decomposition between baseline volume (turn volume) and incremental volume: Within incremental volume, define what portion is attributed to cannibalization with other products, pantry loading (pre- and post-effect), market growth, and competitive brand switching
Create pre-event simulations: Perform what-if evaluations against profitability, quota, and allocated budget
Oracle Demantra Trade Promotion Optimization evaluates past events, business-specific objectives, and constraints and automatically selects the promotion strategy that will produce the best result. It develops and maintains modeling coefficients across the various customers and products using historical inputs, attributes, and other relevant data.
Features
Optimize promotions based on goals, for example, maximize profit/revenue/volume, and constraints, for example, maximum budget or minimum retailer margin.
Evaluate effects of pricing, promotion type, and promotion cost via manufacturer and retailer profitability reports
Create and analyze what-if scenarios: Compare alternative promotion strategies to maximize return on investment
Oracle Demantra Deductions and Settlement Management is the final step in the trade promotion management process. You can track and reconcile invoice deductions and other settlement methods back to promotional events.
Features
Collaboratively manage the settlement resolution process.
Automatically or manually match deductions and settlement methods with promotional events.
Reconcile deductions and payments with proof of performance from retailers based on promotion terms.
Manage reconciliation of trade and non-trade settlements,
Create status reports of deductions clearing process, for example, Settlement Days Outstanding; Days Outstanding by Account, and Settlement Type.
Create detailed ad-hoc reports and high-level metrics for specific business needs.
Receive, research, and clear:
Claims
Deductions
Off-invoice settlements