INVENTORY CASE STUDY Introduction Optimized inventory levels in stores can have a major impact on chain profitability: minimize out-of-stocks reduce overall inventory carrying costs Value chain example Value chain What is the primary objective of most analytic decision support systems ? monitor the performance results of key business processes each business process produces unique metrics at unique time intervals with unique granularity and dimensionality each process typically spawns one or more fact tables value chain provides high-level insight into the overall enterprisedata warehouse Some Common Questions related to Inventory How did the inventory level changed per product, per warehouse over time? How is the profitability of products in our inventory? How many times have we placed a product into an inventory bin on the same day we picked the product from the same bin at a different time? How many separate shipments did we receive from a given vendor, and when did we get them? On which products have we had more than one round of inspection failures that caused return of the product to the vendor? … etc. BI helps answering these questions BI Inventory Models The three main models discussed: Inventory Periodic Snapshot Inventory Transactions Inventory Accumulating Snapshot They are complementary models, and provide different information about the Inventory Periodic Snapshot The most common inventory scheme Example of Retail Store Chain Inventory: The assumed atomic level of detail is: Inventory per product Per day Per Store Basic dimensions: Product Day Store Fact: Inventory Simple Inventory Periodic Snapshot Usage: Provide information about inventory levels: 1. Daily Inventory level 2. Average Inventory level over a time period Problems: 1. Inventory levels are semi-additive (i.e. NOT additive through each dimension) Through the Date dimension the quantity on hand is NOT additive 2. Historical Inventory data using daily granularity results in unreasonably huge amount of data over time Suggestion to define distinct atomic time period for short and long term measures Enhanced Inventory Periodic Snapshot Extra recorded facts Velocity of inventory movement becomes measurable Key concepts: Number of Turns Number of days’ supply Growth Margin Return on Inventory (GMROI) Enhanced Inventory Periodic Snapshot Extra recorded facts measure Number of Turns Number of days’ supply GMROI daily quantity sold quantity on hand quantity on hand quantity sold Over a period total quantity sold dayily average quantity on hand final quantity on hand average quantity sold totalquantitysold x (valueat latest selling price - valueat cost)quantity on hand daily average quantity on hand x value at the latest selling price Enhanced Inventory Periodic Snapshot GMROI - Growth Margin Return on Inventory totalquantitysold x (valueat latest selling price - valueat cost)quantity on hand daily average quantity on hand x value at the latest selling price Turns Gross margin High GMROI lots of turns high gross margin Low GMROI low turns low gross margin GMROI is a standard metric used by inventory analysts to judge a company’s quality of investment in its inventory. We do not store GMROI in the fact table because it is not additive!!! Inventory Transactions Record every transaction that affects inventory: Receive product Place product into inspection hold Release product from inspection hold Return product to vendor due to inspection failure Place product in bin Authorize product for sale Pick product from bin Package product for shipment Ship product to customer Receive product from customer Return product to inventory from customer return Remove product from inventory Inventory Transactions Use: Measure the frequency and timing of specific transaction types Example: • How many times have we placed a product into an inventory bin on the same day we picked the product from the same bin at a different time? • How many separate shipments did we receive from a given vendor, and when did we get them? • On which products have we had more than one round of inspection failures that caused return of the product to the vendor? Inventory Accumulating Snapshot In progress!!! In a single fact table row we track the disposition of the product shipment until it has left the warehouse only possible if we can reliably distinguish products delivered in one shipment from those delivered at a later time also appropriate if we are tracking disposition at very detailed levels, such as by product serial number or lot number Inventory Accumulating Snapshot Fact Table Type Comparison Periodic Snapshot Transaction Accumulating Snapshot Time period represented Regular predictable intervals Point in time Indeterminate time span, typically short lived Grain One row per period One row per transaction event One row per life Table loads Insert Insert Insert and update Row updates Not revisited Not revisited Revisited whenever activity Date dimension End-of-period Transaction date Multiple dates for standard milestones Facts Performance for predefined time interval Transaction activity Performance over finite time General Notes Analysis services presentation: P148-149 – semi additive facts Notes CH6 Aggregate Functions Types in Analysis Services 2008: The functions can be used as a out-of-the-box functions in Analysis services in creating measures Additive 1. a) b) SUM COUNT Pseudo-additive 2. a) b) MIN MAX Non-additive 3. a) b) DistinctCount None Semi-additive 4. a) b) c) d) e) f) FirstChild LastChild FirstNonEmpty LastNonEmpty AverageOfChildren ByAccount Additive, Pseudo- and Non-Additive Aggregate Functions Aggregate Function Category SUM Additive Description Cube demo Data Source specifications The Data Source wizard Cube demo Data Source View specifications The Data Source View wizard Cube demo Implementing named calculations Creating a user friendly date Cube demo Implementing named queries Cube demo Creating dimensions with the Dimension wizard