Presentation to Company ...


Published on

  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Presentation to Company ...

  1. 1. What is Missing to Enable Optimization of Inventory Deployment and Supply Planning? Professor Sridhar Tayur Carnegie Mellon University
  2. 2. ANALYTICS FOR A COHERENT ORDER FULFILLMENT STRATEGY Availability management Key policy choices <ul><li>Promising and meeting order fulfillment lead times </li></ul><ul><li>Set to maintain or gain market share </li></ul>Capacity management <ul><li>Stabilizing production rate to maximize efficiency or flexing capacity to meet demand </li></ul>Demand management <ul><li>Managing sales/order rate variation </li></ul><ul><li>Limiting number of allowed “standard” configurations in build-to-stock environment </li></ul>Inventory management <ul><li>Optimal deployment of inventory to maximize availability at minimum cost </li></ul><ul><li>Also used to insulate manufacturing from demand variability </li></ul>Lead time management <ul><li>Consistent with Lean principles - working to reduce supply and in-process lead-times </li></ul><ul><li>Monitoring and managing lead-time variability </li></ul><ul><li>Fixed or flexible </li></ul><ul><li>Segmentation by product or customer (e.g. sales vs. rentals) </li></ul><ul><li>Fixed or flexible capacity </li></ul><ul><li>Willingness to subject plant to increased demand variability </li></ul><ul><li>Static or dynamic inventory targets </li></ul><ul><li>Rules of thumb vs. product/location/time specific targets </li></ul><ul><li>Based on total chain or local viewpoint </li></ul>To achieve maximum availability at minimum cost: <ul><li>A comprehensive order fulfillment strategy must appropriately define a coordinated set of policies for these interrelated variables </li></ul><ul><li>No one variable can be managed in isolation and changing or fixing one variable has implications for the others </li></ul><ul><li>Active management of demand variability (e.g. promotions/incentives) </li></ul><ul><li>Monitoring and managing forecast error </li></ul><ul><li>Active management of lead-times and lead-time variability </li></ul><ul><li>Incentives and penalties for performance </li></ul>©2002 SmartOps Corporation
  3. 3. ACADEMIC BUILDING BLOCKS: 40+ YEARS OF EVOLUTION, BREAKTHROUGHS, AND APPLICATION <ul><li>Fundamental issues identified setting the stage for decades of research </li></ul><ul><li>Early inventory and stochastic* optimization models created </li></ul><ul><li>Breaking of problems into manageable pieces </li></ul><ul><li>Practitioners use rules of thumb and put pieces together heuristically </li></ul>Late 1950s – 1960s 1970s-1980s 1990s <ul><li>Searching for simpler ways of computing optimal inventory policies for basic problems </li></ul><ul><li>Improved computational approaches developed to address larger problems in “isolation” </li></ul><ul><li>Stochastic optimization models developed to explicitly accommodate supply and demand variability, multiple time periods, capacitated, multi-echelon supply chains </li></ul><ul><li>Successful “one-off” application to industrial-size problems </li></ul><ul><li>Clark and Scarf </li></ul><ul><li>Arrow, Karlin </li></ul><ul><li>Federgruen;Zipkin; </li></ul><ul><li>Lee; Cohen; Roundy </li></ul><ul><li>Muckstadt;Thomas;Zheng </li></ul><ul><li>Glasserman; Tayur </li></ul>Key progress Key contributors ©2002 SmartOps Corporation * Stochastic: Involving or containing random or “uncertain” variables (e.g., uncertain demand, lead time, capacity, yield, etc.)
  4. 4. REAL WORLD: THERE IS SIGNIFICANT INEFFICIENCY IN OUR ECONOMY <ul><li>Fundamental, persistent forces behind supply chain inefficiency: </li></ul><ul><ul><li>Inability to accommodate and actively manage inherent uncertainty, variability, and complexity across multi-echelon supply chains </li></ul></ul><ul><ul><li>Local vs. global (“total cost”) optimization, metrics, and incentives – uncoordinated supply chain inventory and cost decisions within enterprises and across supply chains </li></ul></ul><ul><ul><li>Underutilization of current data, systems, and available best practices, e.g., lack of dynamic, data driven reviews of “planner variability” </li></ul></ul>$1.0 trillion 50+% $500+ billion U.S. inventories Estimated inefficiency Economic opportunity What is missing? What is Missing? Advanced, practical value chain planning and optimization to accommodate and manage these forces ©2002 SmartOps Corporation
  5. 6. CASE STUDY #1: INVENTORY REDUCTION OPPORTUNITY Average inventory (2000) Actual reduction in 2001 Average inventory (2001) Planned reduction in 2002 Average inventory target (2002) Additional opportunity identified with SmartOps Suggested average inventory target (2002) $ Millions Source: SmartOps Multistage Inventory Planning and Optimization Software
  6. 7. CASE STUDY #1: TYPE OF INVENTORY FOR FY2002: ONE PRODUCT LINE AT 95% SERVICE LEVEL $ <ul><li>Key Takeaways </li></ul><ul><li>The existing supply and demand variability drives the need for significant safety stock for products, particularly during the peak selling season </li></ul><ul><li>Due to capacity constraints, there is also a need for pre-build inventory, meaning that plants will produce more inventory not because of system uncertainty, but because mean weekly plant capacity will exceed needed production in future periods </li></ul>
  7. 8. UNDERSTANDING MODELING APPROACHES Annually/quarterly Weekly/daily Quarterly/monthly Low detail/granularity High detail/granularity N/A N/A Planner Planner & O.R. engineer O.R. engineer Business Unit Planning and Operations Corporate/ Business Unit Strategy Organization Data management/update process Relation to existing processes Stand-alone Dynamic One-off studies Driving execution Structural changes Continuous improvement “ Dynamic value chain” ©2002 SmartOps Corporation ERP/APS detailed, dynamic data inputs Manual, “meta-level” inputs, click and drag design SWEET SPOT The goal is to pick an approach that ensures confidence in the answer, quick hit improvements, and sustained execution Timed, regular data loading Data-loader with manual start Data wizard and interface Timing/dynamic frequency
  8. 9. WHAT IS THE OPTIMAL INVENTORY DEPLOYMENT FOR YOUR BUSINESS? Inventory Forms Inventory Purposes To enable continuous and sustained improvement, a comprehensive approach must accommodate all forms and purposes of inventory ©2002 SmartOps Corporation
  9. 10. STOCHASTIC OPTIMIZATION IS NECESSARY <ul><li>Total Cost Optimization </li></ul><ul><ul><li>Cycle stock </li></ul></ul><ul><ul><li>Pre-build stock </li></ul></ul><ul><ul><li>Pipeline stock </li></ul></ul><ul><li>APS challenges </li></ul><ul><ul><li>Scheduling a factory </li></ul></ul><ul><ul><li>Packing a truck </li></ul></ul><ul><ul><li>Routing a truck </li></ul></ul><ul><li>Managing uncertainty </li></ul><ul><ul><li>Safety stock </li></ul></ul><ul><ul><li>Shortfall stock </li></ul></ul>Certain or near-certain “ Deterministic” Uncertain “ Stochastic” Linear and Integer Non-linear Linear, deterministic models are not appropriate for most critical inventory decisions in multistage, multi-product, capacitated, stochastic environments
  10. 11. A SUPPLY CHAIN MODELING PROCESS Map the current value chain Select relevant variables, constraints, and objective function Initial collection, cleaning, and QA of data Selection of planning granularity Select optimization algorithms Commence data integration process <ul><li>Full, partial, or no automation of inputs and outputs </li></ul><ul><li>Entire network or subset </li></ul><ul><li>All nodes or simplification of nodes </li></ul><ul><li>Simplifying assumptions to include or exclude variables, constraints, or nodes considering quality of answer vs. speed of answer </li></ul><ul><li>Understand underlying data assumptions </li></ul><ul><li>Ensure data makes sense in business and supply chain terms </li></ul><ul><li>Days, weeks, months </li></ul><ul><li>Product hierarchy – sales model vs. MA </li></ul><ul><li># of nodes and time periods </li></ul><ul><li>Stationary or non-stationary model (e.g. # of forecast periods) </li></ul><ul><li>Single or multi-echelon or hybrid </li></ul><ul><li>Capacitated, un-capacitated </li></ul>Load data and pre-process meta-data Calculation/ optimization Scenarios/ what-if QA outputs Post-process and summarize Review outputs - send to operational system/ process Change structure of value chain <ul><li>Run test cases vs. actual data </li></ul><ul><li>Understand processing speed </li></ul><ul><li>Design, build, and run logical scenarios </li></ul><ul><li>Test boundary conditions </li></ul><ul><li>Compare results with expectations based on theory and domain expertise </li></ul><ul><li>Aggregation/dis-aggregation </li></ul><ul><li>Units/$s/Weeks </li></ul><ul><li>Rounding </li></ul><ul><li>Manual, exception-based, or automatic export of targets to planning systems </li></ul><ul><li>Changes to “nodes” and “arcs” vs. changes to echelons and BOMs </li></ul><ul><li>Compute meta-data: lead-times, lead time variabilites, forecast disagg. etc. </li></ul>Refresh inputs ©2002 SmartOps Corporation
  12. 13. OVERCOMING PRACTICAL DIFFICULTIES Reality Possible Approach <ul><li>Scale </li></ul><ul><li>Scope: Many Factors Exist Simultaneously </li></ul><ul><li>Data: Existence, Accuracy, Ease of Availability </li></ul><ul><li>Silos within Organizations </li></ul><ul><li>Multiple Companies in a Supply Chain </li></ul><ul><li>Current IT Infrastructure </li></ul><ul><li>Existing Execution and Decision Support Tools </li></ul><ul><li>Metrics and Measurements </li></ul><ul><li>Motivation, Discipline and Incentives </li></ul><ul><li>Training and Capability </li></ul><ul><li>People: Corporate supply chain and business planners/super users as well as business unit planners </li></ul><ul><li>Consultants: Internal and External </li></ul><ul><li>Professors and Education </li></ul><ul><li>Exception Driven Scalable Software </li></ul><ul><li>Comprehensive Approach </li></ul><ul><li>Pre-processors, Inheritors, Data Loaders </li></ul><ul><li>Net Landed Cost View </li></ul><ul><li>Collaborative Framework with Trust </li></ul><ul><li>‘ Bolt-on’s to co-ordinate/synchronize </li></ul><ul><li>Productize recent OR/MS Intellectual Property </li></ul><ul><li>Management 101: Track Key Performance Indicators Dynamically </li></ul><ul><li>Culture and Metrics/Bonus Structure </li></ul><ul><li>Need to have a Grassroots Revolution </li></ul><ul><li>Flexible platform for Multi-tier use and communication </li></ul><ul><li>Do not rely entirely on Spreadsheet based Optimization! </li></ul><ul><li>Appreciate Reality and Train Students to Handle Reality </li></ul>
  13. 14. CLOSING REMARKS <ul><li>Despite ERP and APS investments significant inventory inefficiencies persist </li></ul><ul><li>Fundamental causes of supply chain inefficiency must be addressed: </li></ul><ul><ul><li>Inherent uncertainty and complexity in multistage supply chains </li></ul></ul><ul><ul><ul><li>Stochastic optimization approach is the appropriate solution </li></ul></ul></ul><ul><ul><li>Uncoordinated planning decisions </li></ul></ul><ul><ul><ul><li>Total cost optimization by providing visibility and coordination between functional and external groups </li></ul></ul></ul><ul><ul><li>Inconsistent and/or insufficient planning practices </li></ul></ul><ul><ul><ul><li>Software can provide a standardized “best planning” solution </li></ul></ul></ul><ul><li>All the drivers of inventory must be measured to determine: </li></ul><ul><ul><li>Optimal inventory targets for all inventory purposes </li></ul></ul><ul><ul><ul><li>safety, cycle, shortfall, pipeline, pre-build, and merchandising stock </li></ul></ul></ul><ul><ul><li>Total cost solution to deliver service levels </li></ul></ul><ul><ul><li>Optimal service levels given budget objectives, product margins, and portfolio of products </li></ul></ul>