| Oracle® Retail Advanced Inventory Planning Store and Warehouse Replenishment Planning User Guide for the RPAS Fusion Client Release 14.1 |
|
 Previous |
 Next |
| Oracle® Retail Advanced Inventory Planning Store and Warehouse Replenishment Planning User Guide for the RPAS Fusion Client Release 14.1 |
|
Previous |
Next |
Oracle Retail Advanced Inventory Planning (AIP) is a suite of modules that are designed to manage the supply chain needs of large retailers. These supply chain needs range from interaction with their suppliers through various layers of warehouses down to individual stores and e-commerce sites. It couples time-phased replenishment and allocation algorithms to produce an actionable receipt plan over time. This is based on demand forecasts, replenishment parameters, and inventory availability at the numerous supply points within the supply chain.
AIP takes its place as one of several integrated applications within the Oracle Retail Suite. The suite allows a retailer to manage its supply chain from demand forecasting through to the generation of orders, which can then be shared with collaborative planning partners.
The larger process across the Oracle Retail Suites takes the following form:
Oracle Retail Demand Forecasting (RDF) provides a forecast of consumer demand. This is made available to AIP.
The AIP batch run produces an actionable receipt plan using replenishment parameters maintained inside AIP. Hierarchy and Inventory data is provided by a merchandising system such as RMS.
The receipt plan is then sent to the Order Management module within AIP where those orders that need to be actioned are formally prepared for execution. This includes the assignment of an order number.
Order Management then submits the appropriate orders to the merchandising system where purchase orders and transfers are communicated to other systems. These orders are returned to AIP in subsequent batch runs as orders for on-orders.
Sales forecasts and order plans can then be shared at the appropriate level with suppliers by using a CPFR product so that trading partners can prepare for the forthcoming orders.
At the core of the AIP batch process are five replenishment subprocesses. These subprocesses perform calculations on a set of loaded static and dynamic data using replenishment parameters to produce a replenishment receipt plan for all locations in the supply chain. This functionality differs from many existing replenishment solutions because it yields an actionable plan across time. This gives retailers the ability to project their demand at all levels of the supply chain and to share these projections with their suppliers whenever it is appropriate.
Key to producing an actionable receipt plan is the inclusion of known inventory constraints. This is achieved by performing the five replenishment subprocesses across all locations in the following order:
The Replenishment process generates an Unconstrained Receipt Plan for the part of the planning horizon over which constraints are subsequently applied. This part of the planning horizon is called the Fixed Period.
The Shortfall Reconciliation process applies inventory limitations over the fixed period, modifying the receipt plan where necessary.
The Substitution process attempts to use product substitutes to satisfy any shortage created by Shortfall Reconciliation within the fixed period.
The Stockless process increases orders within the fixed period where necessary to ensure that no warehouse is left with inventory of a stockless product and contractual obligations to order specific quantities of products from vendors are honored.
Finally, the steps are repeated for a second time for the remaining part of the Planning Horizon after the fixed period.
The resulting receipt plan is then exported to the Order Management module where Order Numbers are produced and the orders released to external systems. Following formal order generation, these quantities are fed back into the system and the plan is updated to account for these orders as expected receipts. This type of planning allows the retailer to identify potential supply chain problems before they arise so that stock-outs and excess inventory problems can be prevented or reduced.
The volume of the receipt plan produced by AIP is very large, so a minimum of user involvement is desired, but it must also be possible for super-users to change parameters and to see the effects of their changes quickly in order to avoid persistent costly supply chain problems. The system intends that these parameters are managed principally by exception. Manage by exception means that not every SKU or warehouse is reviewed or updated on a regular basis. Items with exceptions—alerts— are reviewed and updated if necessary.
The user interacts with the AIP system through a number of modules:
Store Replenishment Planning (SRP) workbooks are used to maintain the replenishment characteristics for stores. These workbooks allow the user to analyze system output and perform What-if style analysis when replenishment parameters are changed.
Warehouse Replenishment Planning (WRP) workbooks are used to maintain the replenishment characteristics for warehouses. These workbooks allow the user to analyze system output and perform What-if style analysis when replenishment parameters are changed.
Data Management (DM) is used to maintain the supply chain and network flow information. This includes sourcing links and lead-times.
Order Management (OM) gives the user visibility of order forecasts and previously-executed orders. It permits a range of modifications to previously executed, but not yet received, purchase orders.
The batch replenishment and workbook modules within the AIP Solution run from a common platform, which is called the Oracle Retail Predictive Application Server (RPAS). RPAS is a foundation that includes features, such as:
Multidimensional databases
Product, time, and business location hierarchies
Aggregation and spreading of data
Workbooks and views for displaying and manipulating forecast data
Wizards for creating and formatting workbooks and views
Menus, quick menus, and toolbars for working with forecast and sales data
Exception management and user-friendly alerts
See the Oracle Retail Predictive Application Server User Guide for the RPAS Fusion Client for more details.
This section describes the basics of the RPAS user interface.
A workbook is an easily viewed, easily manipulated multidimensional framework that is used to perform specific business functions, such as generating replenishment receipt plans and reviewing parameters. To present data, a workbook can contain any number of multidimensional spreadsheets, called views, as well as graphical charts. These components work together to facilitate viewing and analysis of business functions.
Data in a workbook can be viewed at lower levels of detail or higher levels of aggregation. Different views are obtained by changing the path, or level of data rollup, or both. Data in a workbook can also be manipulated at any hierarchical level. If you modify data at an aggregate level, these changes are distributed down to the lower levels. The reverse is also true— if you modify data at a lower level in the hierarchy, the aggregates of the data reflect those changes.
Views are multidimensional spreadsheets that are used to display workbook-specific information. Workbooks can include one or many views. Views can present data in the form of numbers in a grid, or these numeric data values can easily be converted to a graphical chart.
You can display the information in a view in a variety of formats, generally by rotating, pivoting, and changing the data rollup. These processes are explained in detail in the Oracle Retail Predictive Application Server Online Help for the RPAS Fusion Client, which is available from the Help menus in SRP/WRP or the accompanying print version of the Oracle Retail Predictive Application Server User Guide for the RPAS Fusion Client. You can easily change the presentation style of data in a view. When data sets are moved around, their positions change. Data can be viewed at a very high level of detail, or data values can be quickly aggregated and viewed at summary levels. Furthermore, views and parts of views can easily be changed into charts, which facilitate the graphical viewing and analysis of data.
When new workbooks are created, assistance in defining the nature of the information displayed is provided in the form of wizards. Wizards are pre-defined steps that walk the user through the data selections necessary to use/build each workbook. Wizards are also available to prompt the user for information regarding the definition of alerts, and user account management.
This section describes:
The SRP workbooks allow you to interact with the AIP replenishment system from a store perspective and the WRP workbooks allow you to interact from a warehouse perspective. Neither SRP nor WRP exists in its own right as a batch process.
Instead, the actual process of producing a store or warehouse replenishment plan is accomplished by the running of the batch Replenishment steps.
These steps are process oriented, rather than being location specific. Instead of each step being focused on just stores or just warehouses, each one is performed for all appropriate nodes in the network.
When strung together, these batch steps produce a Constrained Receipt Plan for all locations in the network. SRP provides a store-oriented mechanism for interacting with the receipt plan while WRP provides a warehouse-oriented mechanism for interacting with the receipt plan.
AIP supports a basic business process for creating and releasing replenishment plans for all locations in the network. This process combines batch and interactive online activities as shown in Figure 2-2.
To generate a constrained replenishment plan for any location, four main types of data must first be loaded into the RPAS database:
Basic hierarchy and attribute information
Forecast Store Sales
Inventory information, including on-hand and on-order information that has previously been created by AIP
Supply chain definition data including release schedules, and sourcing links
Replenishment is run for the fixed period. The fixed period refers to the first part of the Planning Horizon in which inventory limitations apply. Replenishment is a destination-centric process. It addresses every node in the supply chain, starting with stores and progressing onto warehouses, identifying the unconstrained orders required within the fixed period for every SKU Pack/destination combination.
The orders for each SKU Pack/destination combination are calculated over the fixed period using the latest inventory position, on-orders and in-transit information. Forecast sales represent demand on a store. For a warehouse however, the demand stream is determined by aggregating the orders of the stores served by the warehouse. These store orders are the ones created by this process step. This means that replenishment for stores is performed before replenishment for warehouses.
For each destination, the available information is used to make projections of future inventory positions for each delivery opportunity. If the projection for a delivery opportunity falls below boundaries dictated by the replenishment method, then an order is required. Each order is then subject to rounding.
The resulting orders represent the Unconstrained Receipt Plan within the fixed period.
Shortfall reconciliation is source-centric and therefore works from the top to the bottom through the nodes in the supply chain network. The inventory available at the source is evaluated for each reconciliation period and is distributed against the need of destinations that ship with the particular period. Where a destination has need of a relational Event SKU (promotion with variant, replacement SKU) the related SKU inventory is also assessed and substituted where needed.
The goal of the process is to meet each destinations need in the most equitable way, according to priority. The Watershed algorithm is applied to a configured priority matrix—which defines a SKU's priority at a store—in an attempt to give destinations of the same priority an equal chance of meeting the demand placed upon them.
When a shortage occurs at a store, the unmet demand may impact similar SKUs by increasing demand as shoppers choose a different available flavor, color/style, or correlated product. The SKU impact analysis is performed outside of AIP and a resulting Demand Transfer Percentage (DTP) is given to AIP. The substitution process identifies the times when the store inventory falls short of demand and uses the DTP to assess the increased demand on substitute SKUs against the inventory available. Where there is not enough excess in the store an attempt is made to send more. The same Watershed algorithm used in shortfall reconciliation is used to share additional inventory amongst store destinations experiencing a shortage.
Stockless functionality identifies those situations where there is excess stock at a Source that needs to be consumed by the destinations served by the source. Stockless is a source-centric process and works for vendors with Supplier Purchase Quantity (SPQ) and warehouses with stockless products.
Stockless identifies excess stock at sources. Excess stock is defined as any remaining inventory of a stockless product at a warehouse not consumed by demand, or any residual inventory at a Vendor that the retailer is contractually committed to the order.
Stockless attempts to increase orders from the source to absorb the excess. The priority matrix, watershed and fairshare algorithms are used to ensure that destinations are fairly dealt with.
The resulting orders represent the final Constrained Receipt Plan for the fixed period. The orders in the fixed period are not changed any further within the current batch run.
A second run of Replenishment is performed but this time focusing on the remaining planning horizon after the fixed period. This replenishment run is based on the inventory position resulting from fixed period reconciliation, substitution, and stockless pushing.
For those SKUs experiencing a significant supply chain event or are stockless Shortfall Reconciliation and Stockless processing is also performed in the post fixed period. The events triggering post fixed period reconciliation for a SKU are:
Stockless at a Source
Temporarily Un-orderable
Promotion with Variant Event
Replacement SKU Event
Pack Change Event
Discontinuation date set of a SKU's ordered pack
Stop receiving date set of a SKU's ordered pack
The final resulting receipt plans from all fixed period and post fixed period processing are referred to as the Constrained Receipt Plan for the planning horizon. It is this receipt plan that can be viewed in the SRP and WRP reporting and analyzing activities.
Once a Constrained Receipt Plan has been produced, a subset of the orders is then sent to the Oracle Platform for processing by Order Management. This subset contains the following:
All into store and into warehouse orders sourced from Vendors (Purchase Orders) across the entire planning horizon.
All into warehouse orders sourced from other warehouses (Transfers) across the entire planning horizon.
All into store orders sourced from warehouses (Transfers) with release dates that need to be shipped today.
Of the subset of orders exported to Order Management, those with a release date of today are given appropriate orders numbers. The Order Numbers are allotted based upon an implementation-time mask that determines how orders (both Transfers and Purchase Orders) should be grouped together.
Those orders that have been allotted numbers are then executed (released) to the merchandising system for subsequent communication to other systems including EDI.
The Order Management interface gives the user access to those orders that have already been released. Those purchase orders that have not been received may be updated. Those forecast Purchase Orders that have not yet been released because they ship at a later date may be executed early.
