Demand Planning Overview
This chapter covers the following topics:
About Oracle Demand Planning
Oracle Demand Planning is a web-based application that enables organizations to produce unconstrained forecasts for future demand and generate tactical, operational, and strategic business plans. Demand Planning captures and processes information from multiple sources and consolidates demand so that it can be summarized by item, product line, region, time, and organization. Demand Planning uses Oracle Workflow and supports control mechanisms based on an event or calendar.
Oracle Demand Planning is integrated with Oracle Inventory Optimization (IO), Oracle Collaborative Planning (CP), and Oracle Advanced Supply Chain Planning (ASCP) to ensure optimal inventory planning and supply chain management.
Key features of Oracle Demand Planning
The following list highlights features of Oracle Demand Planning:
-
Supports Internet-based collaboration between users in various locations and departments. The collaborative capabilities ensure global visibility and provide tools for adjusting forecasts based on judgment.
-
Incorporates a best-of-breed statistical forecasting engine: Geneva ForecastingTM from Roadmap Technologies. Developed specifically for demand forecasting applications, this engine has multiple forecasting methods (linear and non-linear regression, variations of exponential smoothing), efficient parameter search and outlier detection algorithms, and a rule based system for automatic best fit forecasts based on time and frequency domain analysis and predefined decision rules generated from cross-client studies. The statistical interface supports both novice and expert users.
-
Accommodates flexibility in information acquisition, information management, information analysis, and forecast reconciliation. The system supports a variety of inputs including sales and opportunities forecasts, marketing intelligence, custom forecasts, and external data streams.
-
Enables end users to define attributes of custom data streams such as sales forecasts, measurement type, aggregation rule, allocation rule, unit of measure, dependent demand, price list, and related events.
-
Provides a reporting interface that supports multiple views of the data, rotation, drill down, and aggregation. The ability for planners to define their own measures, custom aggregates, and saved selections provides further support for flexible modeling and analysis.
-
Enables planners and the Demand Plan Manager to use a variety of techniques to edit data.
-
Creates a knowledge base by maintaining planners' comments and audit trails. Supports reason codes for forecast modifications.
-
Includes capabilities for modeling and viewing information about events such as product introductions, product cannibalization, and product phase outs.
-
Provides feedback to planners through performance monitoring, exception reporting, comparative reports, and user-defined alert mechanisms.
Demand Planning Architecture
Oracle Demand Planning has component architecture. This means that in lieu of working directly from data generated by other Oracle Applications, Oracle Demand Planning copies the data to a localized data store in a process called collection. This is done for two reasons:
-
It allows the computational-intensive planning and forecasting calculations to be off-loaded to a separate Demand Planning Server, where the localized data store resides, to avoid excessive load on the server hosting the transactions.
-
It allows Oracle Demand Planning to be deployed against multiple versions of Oracle Applications (versions 10.7, 11, 11iand 12). There are separate collection programs for each version.
The business process of demand planning requires analyzing demand along different dimension levels. For example, manufacturing planners may plan and view demand at a product/manufacturing location level, whereas sales managers may plan and view demand at a product family/geographic region level. The ability to accept demand data input at one set of dimension levels, display it at another set of dimensional levels, and maintain a consistent set of underlying demand data that is independent of the levels at which it is displayed is covered by a technique known as Online Analytical Processing (OLAP). The Oracle 9i database release 2 contains Online Analytical Process.
A representation of Oracle Demand Planning in the Oracle Database is illustrated below shows that Relational and Online Analytic Process data are managed in a single, integrated instance. This integration allows for an internal movement of data when downloading data from the Demand Planning Server to the Demand Planning Engine and uploading data from the Demand Planning Engine to the Demand Planning Server.
The Oracle Database stores multidimensional data directly in the database. The database engine that runs the online analytical process calculations for Oracle Demand Planning is the OLAP option in the database known as Analytic Workspaces. The database engine runs within the database kernel and internally executes all multidimensional data calculations. Analytic Workspaces contain objects such as dimensions, variables, formulas, and OLAP Data Manipulation Language (DML).
All data in the Oracle Database, both relational and multidimensional, is stored in Oracle data files. Multidimensional data is stored in an Analytic Workspace within a relational schema. An Analytic Workspace is a container for collections of multidimensional data types and may contain one or many 'cubes'. These cubes are the plans created in Oracle Demand Planning.
Oracle Demand Planning consists of two major components: the Demand Planning Server and the Demand Planning Engine. The Demand Planning Server holds the inputs such as sales history, as well as the forecast outputs for feeding to Oracle Advanced Supply Chain Planning and other Oracle applications. The Demand Planning Engine is the processing engine of Oracle Demand Planning, and is based on Analytic Workspaces.
Planning Cycle
The following are the steps in the planning cycle.
Collecting data
Data such as items, sales history, and inventory organizations are collected from the source into staging tables in the Demand Planning Server. The purpose of the staging tables is to provide a temporary repository that allows users to review the collected data, adjust the data as necessary, and clean out any irrelevant data, thus making the data more useful for forecasting. Note that Oracle Demand Planning does not provide an explicit data viewing or cleansing tool. You can use any data manipulation tool such as SQL for this purpose. The data are then pulled into fact tables on the Demand Planning Server. If the data are clean, then it can be collected directly into the fact tables.
Defining the demand plan
The administrator defines a demand plan in the Demand Planning Server. He or she specifies the dimensions and hierarchies within dimensions such as geography and sales group. The administrator also specifies one or more scenarios based on different histories or date ranges for forecast horizons.
The administrator also specifies what type of historical data will be used and how much of the history to use. In addition to historical data, the administrator can select reference data such as manufacturing data for comparing forecasts generated by Oracle Demand Planning based on historical data.
If the system collects data from the Bills of Material (BOM), the administrator can set up the Demand Planning Server so that planners and the Demand Plan Manager can view and edit data for dependent demand and planning percentages.
Downloading data from the Demand Planning Server to the shared database
Data from the Demand Planning Server database is downloaded into a shared database in the Demand Planning Engine. The Demand Planning Server is a source of integrated data that can include the following:
-
Forecast data from Sales, Manufacturing, Supply Chain, or third party at any aggregate level.
-
Historical data at the lowest aggregate level (e.g., SKU).
-
Planning percentages from the Bills of Material (BOM).
-
Data from external sources, such as customer data.
The Demand Planning Server defines the parameters of the source data as well as events, scenarios, demand plans and system setup. The downloaded data is used as a basis for populating measures which hold the information for data streams from the Demand Planning Server such as booking history or customer forecasts. The Demand Planning Administrator can create additional measures to be used planners and the Demand Plan Manager.
Distributing data to planners
Data in the shared database are distributed to individual planners, based on assignments that are defined by the Demand Planning System Administrator. Assignments determine the dimension value combinations (measures and dimension values) that each planner is responsible for forecasting and submitting to the shared database as well as the dimension value combinations that each planner can view. In a collaborative planning environment, assignments establish the areas by which planners can collaborate.
Creating and modifying forecasts
Individual planners generate forecasts and use worksheets to review and modify forecast values. Planners can adjust the data in a number of ways, select different data values, and copy and paste data. As planners work, they can use ad hoc reports and graphs to visualize data, perform what-if analyses, and look at the data in a variety of ways. Planners can also define custom data streams to use in analyses. Planners can share their most recent updates with the Demand Plan Manager and other authorized collaborators.
Selecting final forecasts to submit to the shared database
When forecasting is complete, planners choose the measure that contains the forecast data that they want to submit to the shared database for each scenario in the demand plan.
Reviewing forecasts
The Demand Plan Manager reviews the consolidated forecast in the shared database. He or she might make further judgmental adjustments. Optionally, he or she can upload the forecast to the Demand Planning Server.
Uploading data to the Demand Planning Server
If the Demand Plan Manager has not already performed this function, the Demand Plan Administrator uploads the consolidated forecast from the shared database to the Demand Planning Server.
Publishing final forecasts and renewing the planning cycle
The consolidated forecast is then used for supply chain planning and scheduling. The forecast can be:
-
Included in a supply plan in Oracle Advanced Supply Chain Planning.
-
Included along with the forecast accuracy information in an inventory plan in Oracle Inventory Optimization.
-
Published to Oracle Collaborative Planning.
-
Published from the Demand Planning Server to the source instance to integrate the final estimate of demand into the transaction system.
The same demand plan can also be used for the next planning cycle after adjusting the time horizon.
Demand Planning Roles
Oracle Demand Planning supports the following user roles:
Demand Planning System Administrator
The Demand Planning System Administrator is responsible for integrating Oracle Demand Planning with other products and for administering the planning cycle. The integration requires collecting data into Oracle Demand Planning from the source transaction instance, such as Oracle Applications, making sure that data is clean and usable, and publishing Oracle Demand Planning output back to the source for planning and scheduling.
The setup tasks require determining overall default settings for the demand planning system from a deep understanding of the business process. This includes specifying the forecast level, assigning data to individual demand planners, selecting and setting defaults for predefined reports, and invoking data consolidation after planners send their forecasts to the shared database.
The Demand Planning System Administrator should have thorough knowledge of the demand planning process and be familiar with Demand Planning Analytic Workspaces.
Demand Plan Manager
The Demand Plan Manager gets a view of the entire data without any restriction or scoping. He or she works directly in the shared database and is responsible for reviewing planners' consolidated forecasts.
The Demand Plan Manager is responsible for the final forecast numbers that are written back to the Demand Planning Server. After planners submit their forecasts, the Demand Plan Manager reviews the consolidated forecast numbers and decides whether to accept or reject them. He or she could modify the numbers, or choose to reassign them to the individual planners.
Note: If your planning group is very small, you might want to set up all users to have the Demand Plan Manager responsibility. Note, however that the Demand Plan Manager works directly in the shared database. This means that if more than one person has this responsibility, only one individual at a time will be able to perform functions such as creating measures, editing data, and saving documents.
Demand Planner
The Demand Planner is responsible for analyzing and forecasting demand in an assigned data segment and for submitting demand forecasts. A Demand Planner is an individual forecaster. Qualifications for this role include forecasting and data analysis abilities and a good understanding of the business processes governing the data to be forecast.
Demand Plan Viewer
The Demand Plan Viewer role allows users read-only access to data in the Shared database. Demand plans can be viewed simultaneously by multiple users without creating additional Planner assignments.
When using this role, data cannot be changed, shared or submitted, but documents shared by the Plan Manager or Planners are visible. Reports and graphs can be viewed, exported, printed and saved. In addition, the user can create and share reports, graphs, and saved selections.
The document toolbar and selector are available for use in this role. Alert notifications can be received.