1. University of Minnesota – Digital Technology Center December 2002 Intelligent e-Supply Chain Decision Support Norman M. Sadeh e-Supply Chain Management Laboratory School of Computer Science Carnegie Mellon University

2. Copyright ©2002 Norman Sadeh Outline Supply Chain Management: New Context Supply Chain Management: New Context Agent-Based Collaborative Decision Support Agent-Based Collaborative Decision Support  Mascot Available-To-Promise/Capacity-To-Promise Functionality Available-To-Promise/Capacity-To-Promise Functionality Empirical Results Empirical Results Dynamic Supply Chain Management Practices Dynamic Supply Chain Management Practices  Early Results  TAC’03: A Supply Chain Trading Competition Summary and Concluding Remarks Summary and Concluding Remarks

3. Copyright ©2002 Norman Sadeh Supply Chain Management Planning and coordinating procurement, production and distribution activities Planning and coordinating procurement, production and distribution activities  From raw material suppliers to manufacturers …to distribution centers …to retailers and consumers Trillions of dollars annually Trillions of dollars annually Good practices directly impact the competitiveness of companies Good practices directly impact the competitiveness of companies  Timely and cost-effective delivery of products to customers  Extends to product design and configuration

4. Copyright ©2002 Norman Sadeh Why is SCM Difficult? Involves multiple organizations Involves multiple organizations Each organization tries to satisfy multiple objectives Each organization tries to satisfy multiple objectives  Cost, timeliness, quality, market share, etc. Each organization operates subject to: Each organization operates subject to:  Internal Considerations:  Finite capacity, existing inventory, etc.  External Considerations  Available suppliers and their capacities, order quantities and due dates, contractual arrangements, transportation constraints, etc. Numerous sources of uncertainty Numerous sources of uncertainty  Capacity, supplies, demand, etc.

5. Copyright ©2002 Norman Sadeh Historical Perspective PurchasingProductionSales Distribution Materials Management Manufacturing Management Distribution Customers Suppliers Internal Supply Chain Customers Buyers/ Sellers Buyers/ Sellers Buyers/ Sellers Buyers/ Sellers Inventory Control e-Commerce e-Markets/Exchanges e-Supply Chains Functional Silos Enterprise Integration Dynamic Internet-enabled Supply Chain Integration

6. Copyright ©2002 Norman Sadeh Beyond the Early eMarket Hype Dynamic business practices are mainly confined to MRO Dynamic business practices are mainly confined to MRO Suppliers don’t like being evaluated solely based on price Suppliers don’t like being evaluated solely based on price  Covisint, E2open exchanges: more emphasis on supporting collaboration  Requires richer environments Multiple attributes – not just priceMultiple attributes – not just price  Lack of adequate standards Lack of adequate decision support tools Lack of adequate decision support tools  Evaluate a large number of options Standardization efforts are taking time Standardization efforts are taking time

7. Copyright ©2002 Norman Sadeh Some Open Research Issues Long vs. Short term contracts Long vs. Short term contracts Information exchange Information exchange Collaborative decision support Collaborative decision support Multi-attribute negotiation Multi-attribute negotiation Peer-To-Peer/local view vs. more lobal view Peer-To-Peer/local view vs. more lobal view  P2P Challenge: Coordinating negotiation across multiple tiers  Challenge for the Global View:  Creating the right incentives for information sharing  How global? How often do you clear? etc.

8. Copyright ©2002 Norman Sadeh MASCOT: Collaborative Decision Support  Decisions are evaluated in collaboration with potential business partners  Supply chains can be dynamically set up in response to changing market requirements  Emphasis on Mixed Initiative Decision Support  Don’t try to automate everything!

9. Copyright ©2002 Norman Sadeh MASCOT Supply Chain Agent Bidding & Order Mgmt. Planning & Scheduling Procurement Business Entity Enterprise Level Mascot Agent Tier 1 Suppliers eMarket Site Level Mascot Agent Site Level Mascot Agent Prospective Customer Request for Quote eMarket

10. Copyright ©2002 Norman Sadeh MASCOT: Overall Objectives Leverage benefits of finite scheduling Leverage benefits of finite scheduling Rapid and accurate evaluation of partner-dependent decisions : Rapid and accurate evaluation of partner-dependent decisions :  Bids & Requests for Quotes including real-time ATP/CTPincluding real-time ATP/CTP  Alternative product/subcomponent designs  Make-or-buy decisions Customizable mixed-initiative functionality Customizable mixed-initiative functionality  Collaborative solution development, workflow management Facilitate integration with legacy systems Facilitate integration with legacy systems

11. Copyright ©2002 Norman Sadeh A Customizable Agent Wrapper Blackboard Control KB Agenda Controller Control Profile GUI Event Queue Enterprise System eMarket Portal Potential Business Partner Potential Customer Bidding/Order Mgmt KS BOM/Flow Config. KS Demand Explosion KS Communication KS Knowledge Sources Contexts External Systems Outgoing messages Incoming Events Scheduling/CTP KSRFQ/Procurement KS Current working context

12. Copyright ©2002 Norman Sadeh Main Architectural Features Blackboard Contexts: “What-if” Blackboard Contexts: “What-if”  Different assumptions (e.g. demand, resources, suppliers) and different solutions  Unresolved issues Extensible set of Knowledge Sources (KSs) Extensible set of Knowledge Sources (KSs)  Allows for modular & reusable KSs  Provides for easy integration with legacy systems Mixed Initiative Control Mixed Initiative Control  Customizable user profile

13. Copyright ©2002 Norman Sadeh Unresolved Issues Help keep track of incomplete, inconsistent and unsatisfactory aspects of a context solution Help keep track of incomplete, inconsistent and unsatisfactory aspects of a context solution  Examples: unprocessed RFQ, insufficient availability of supplies, missed prior delivery commitment Automatically updated as the solution is modified Automatically updated as the solution is modified Supports flexible mixed initiative workflow management Supports flexible mixed initiative workflow management  Associated with KS activations, scripts and goals

14. Copyright ©2002 Norman Sadeh Three Levels of Problem Solving Knowledge Source Activations Knowledge Source Activations  e.g. Demand Explosion (RFQ1) Scripts Scripts  e.g. Evaluate (RFQ1) 1. Copy_Current_Context 2. Incorporate(RFQ1) 3. Demand_Explosion (RFQ1) 4. Reoptimize_Schedule_with_Net_Demand (RFQ1) 5. Procure_Subcomponents_Net_Demand(RFQ1) 6. etc. Goals: Search among multiple options

15. Copyright ©2002 Norman Sadeh Mixed Initiative Workflow Management U/C: Incorporate event into Context Blackboard: Update Unresolved Issues U/C: Select Unresolved Issue to Resolve U/C: Select Resolution Method KS: Execute Resolution Method Controller: Activate Resolution Method U/C: Modify Assumptions within Context (“What-if”) Modified Context Modified Context Unresolved Issue Selected Unresolved Issue Selected Resolution Method Activated Agenda Item Modified Copy of Context U/C = User or Controller

16. Copyright ©2002 Norman Sadeh Status Customized to support coordination between a machine shop and a tool shop at Raytheon Customized to support coordination between a machine shop and a tool shop at Raytheon  Over 150 machine centers & over 100 people  50% of incoming orders require new tools  alternative BOM & process planning options  Reduced tardiness by 23 percent  Integration of process planning & scheduling  Tighter coordination Used to study the benefits of different supply chain coordination policies and different order promising policies Used to study the benefits of different supply chain coordination policies and different order promising policies

17. Copyright ©2002 Norman Sadeh Dynamic Supply Chain Coordination orders requests for bid bid acceptance/rejection order cancelation/modification products bid submissions revised delivery dates Supply chain entity Supply chain entity Supply chain entity Supply chain entity

18. Copyright ©2002 Norman Sadeh The Coordination Challenge Generate robust yet competitive and cost-effective promise dates Generate robust yet competitive and cost-effective promise dates  Multi-tier “capacity-to-promise” functionality Sources of uncertainty are both internal and external Sources of uncertainty are both internal and external  incoming orders, supplies, internal capacity, etc. Is it possible, through dynamic coordination, to reap the benefits of finite scheduling, while offsetting the brittleness of its solutions ? Is it possible, through dynamic coordination, to reap the benefits of finite scheduling, while offsetting the brittleness of its solutions ?

19. Copyright ©2002 Norman Sadeh Real-Time Promising(RTP): General Considerations Net Demand: Inventory Allocation & Demand Explosion Net Demand: Inventory Allocation & Demand Explosion Scheduling Scheduling  Available vs. modified capacity  Schedule around prior commitments vs. reoptimization  Schedule Reoptimization  Assess impact on prior commitments  Costs & Priorities: order priorities, late delivery penalties, inventory costs, etc.  Other Tradeoff: Speed versus “optimality” Assess desirability & decide whether to submit quote Assess desirability & decide whether to submit quote Micro-Boss RTP module: real-time reoptimization - user specifies desired response time (Sadeh et al. ‘94-99) Micro-Boss RTP module: real-time reoptimization - user specifies desired response time (Sadeh et al. ‘94-99)

20. Copyright ©2002 Norman Sadeh RTP: Further Refinements Profitable-To-Promise Profitable-To-Promise Selective RTP Validation Selective RTP Validation

21. Copyright ©2002 Norman Sadeh Profitable-To-Promise Overall Profit = Total_Revenue - Total_Costs Overall Profit = Total_Revenue - Total_Costs  Total_Revenue: Sum over all orders  Total_Costs: Production costs, inventory costs (raw materials, in-process, finished goods), late delivery penalties, etc.  Takes into account impact on prior commitments  e.g. late delivery penalty when another order gets bumped Bid only if overall profit increases Bid only if overall profit increases  Other variations can be considered  e.g. strategic customers, market share considerations

22. Copyright ©2002 Norman Sadeh Empirical Study: Multiple RFQ Processing Policies Response: Response:  Always bid - no due date negotiation  Only submit a bid if overall profit increases  Bid conditional on acceptance of possibly relaxed promise date Capacity-To-Promise Computation Capacity-To-Promise Computation  Leadtime-based  Local finite capacity scheduling & supply leadtimes  Coordinated finite capacity scheduling

23. Copyright ©2002 Norman Sadeh Empirical Study: Assumptions  A lot-for-lot make-to-order environment  Internal sources of uncertainty at each tier due to resource breakdowns and variations in processing times  Stochastic order arrival  Finite capacity schedules regenerated daily  Micro-Boss scheduling system  JIT objective: minimize sum of tardiness & inventory costs  Execution priority in accordance with the latest released schedule

24. Copyright ©2002 Norman Sadeh Evaluation Criteria  Number of bids refused or rejected  Number of tardy orders  Average utilization of the most utilized resource  Average supply chain leadtimes  Average due-date adjustment (as part of bid negotiation)  Profit (sales revenue minus costs)  Total in-system inventory costs (WIP and finished goods)  Total tardiness costs  Promise date accuracy

25. Copyright ©2002 Norman Sadeh Basic Supply Chain Configuration Agent 1 Agent 2 Agent 3 Agent 4 Agent 5 Agent 9 Agent 10 Agent 6 Agent 7 Agent 8 External customers Supply chain Tier3Tier2Tier1

26. Copyright ©2002 Norman Sadeh Benefits of Dynamic Finite Capacity Coordination 0 3 6 9 12 15 18 Leadtime-basedLocal FCSCoordinated FCS Average lead time (days) (in bars) -2000000 -1000000 0 1000000 2000000 3000000 4000000 Profit (in high-low-lines) Case with competition and negotiable promise dates

27. Copyright ©2002 Norman Sadeh Benefits of Dynamic Coordination - Contd. -6000000 -4000000 -2000000 0 2000000 4000000 6000000 0,400,600,801,001,20 Nominal load Average profit Leadtime-based Local FCS Coordinated FCS

28. Copyright ©2002 Norman Sadeh Dynamic Supplier Selection A manufacturer has a given set of customer orders to satisfy A manufacturer has a given set of customer orders to satisfy Each order has a required delivery date along with a penalty for missing that date Each order has a required delivery date along with a penalty for missing that date The manufacturer’s capacity is finite The manufacturer’s capacity is finite Each order requires a number of components for which suppliers have submitted bids Each order requires a number of components for which suppliers have submitted bids  Supply bids include a price and delivery date

29. Copyright ©2002 Norman Sadeh Supplier Bid Selection Component 11 Order 1 Order 2 Component 12 Component 21 Component 22 Component 23 Supplier Bid (Price, Delivery Date) (Delivery Date, Late Penalty)

30. Copyright ©2002 Norman Sadeh Trading Agent Competition TAC Classic: Travel Agent Scenarios TAC Classic: Travel Agent Scenarios  About 20 entries in the past TAC’03: Supply Chain Trading Competition TAC’03: Supply Chain Trading Competition  Agents compete for supplies and demand  Fixed Assembly Capacity  RFQs from customers – delivery date and tardiness penalty  RFQs to suppliers  Interests on money borrowed from the bank

31. Copyright ©2002 Norman Sadeh

32. Summary e-SCM is about more open and more dynamic business practices e-SCM is about more open and more dynamic business practices Mascot: Mascot:  Rapid evaluation of partner-dependent decisions  Mixed initiative decision support  Coordinated real-time Profitable-To-Promise functionality Ongoing work: Ongoing work:  Combine e-SCM and multi-attribute negotiation – together with the Univ. of Michigan  Dynamic Supplier Selection  Trading Agent Competition

33. Copyright ©2002 Norman Sadeh Q&A