1. Ford Motor Company’s Finished Vehicle Distribution System April 2001 Ellen Ewing Project Director UPS Logistics Dr. John Vande Vate Exec. Director EMIL ISyE Georgia Tech

2. Confidential Page 2 Confidential Agenda l Introduction l 1999 Environment l Solution Approach l Network Design l Implement New Strategy l Results to Date l Summary

3. Introduction

4. Confidential Page 4 Confidential Objectives/Motivation l Importance of Pipeline Inventory l Pipeline Inventory and Network Design l Role of modeling l Information in variables l Stronger formulation l Financial impact

5. Confidential Page 5 Confidential Competitive Necessity The new BMW Sales and Production System

6. Confidential Page 6 Confidential Financial Incentives: Capital Utilization –In 1996 –Ford produced 3.9 million vehicles in the US –Avg. transit time 15+ days –Avg. vehicle revenue $18,000 –Value of pipeline inventory: > $2.8 Billion –One day reduced transit time: »$190 Million reduction in pipeline inv. »1,400 fewer railcars The Need for Speed

7. Confidential Page 7 Confidential The Need for Speed Demand for land 22 Plants 54 Destination Ramps ~1,200 Load lanes ~8,400 vehicles waiting at plants $166 Million in inventory

8. Confidential Page 8 Confidential The Need for Speed Other Incentives l Damage l Flexibility l Others? l

9. Confidential Page 9 Confidential Before 1996

10. Confidential Page 10 Confidential

11. Confidential Page 11 Confidential The Price Inventory at the cross dock Added distance traveled Handling at the cross dock Capital costs of the cross dock

12. Confidential Page 12 Confidential Mixing Centers

13. 1999 environment

14. Confidential Page 14 Confidential 1999 Vehicle Network Delivery Conditions l Record production levels l Demand shift from cars to trucks l Overburdened rail infrastructure l Deteriorating rail service l Shortage of transport capacity l Mixing centers l 15+ day transit time l High inventory cost l Dissatisfied customers

15. Confidential Page 15 Confidential High 1999 Level Statistics l Assembly plants22 l Mixing centers 5 l Destination rail ramps54 l Dealer locations6,000 l Production volume4.4 Mil./Year l Freight expense$1.5 Bil. l Dec. ‘99 avg. transit time16.8 Days l Pipeline Inventory$4.1 Bil.

16. Confidential Page 16 Confidential Mixing Center Origin Plant Groupings Destination RampPlanned Ramp Closure Edison Norfolk Atlanta Kentucky Ohio St Louis Canada St Paul Michigan Chicago Kansas City 15% of all vehicles go Haulaway Direct to Dealer within 200-300 Miles of the Assembly Plant 85% of all Vehicles go via Rail to a Hub (Mixing Center or Destination Ramp) Ford Distribution Network

17. Confidential Page 17 Confidential Old Delivery Design l Push Network l Vendor sub systems optimized for individual segments l Little to no visibility l Mixing Centers not used effectively

18. solution approach

19. Confidential Page 19 Confidential Ford Goals Speed l 1999: Average 15 days transit time l Goal: Maximum of 8 days transit time Precision l 1998/1999: 37% on time within 1 week l Goal: 95% on time within 1 day Visibility l 100 % Internet vehicle tracking from plant release to dealer delivery l Guide the flow of vehicles l Respond to variations l Inform customers

20. network design

21. Confidential Page 21 Confidential Design Process  Truck vs Rail delivery  Allocate Dealers (FIPS) to Ramps  Route Flows through Rail Network

22. Plant Mixing Center Origin Ramp Dest. Ramp MC Ramp Ford Locations

23. Confidential Page 23 Confidential An Allocation Model

24. Confidential Page 24 Confidential Single-Sourcing Allocation Var Assign{FIPS, RAMPS} binary; Minimize TotalCost: sum{fip in FIPS,ramp in RAMPS} Cost[fip,ramp]*Assign[fip,ramp]; s.t. SingleSource{fip in FIPS}: sum{ramp in RAMPS}Assign[fip,ramp] = 1; s.t. ObserveCapacity{ramp in RAMPS}: sum{fip in FIPS} Volume[fip]*Assign[fip,ramp] <= Capacity[ramp];

25. Confidential Page 25 Confidential Dealers sourced by multiple ramps Old Ramp Allocation Southern US

26. Confidential Page 26 Confidential Dealers sourced by single ramps New Ramp Allocation Southern US

27. Confidential Page 27 Confidential New Allocation of Dealers to Ramps Mainland US

28. Confidential Page 28 Confidential Flows through the Rail Network Objective is NOT Freight cost!

29. Confidential Page 29 Confidential The Objective IS Speed Capital Land

30. Confidential Page 30 Confidential The Promise Speed Unit trains bypass hump yards

31. Confidential Page 31 Confidential The Promise Capital & Land Time Laurel, Montana Orilla, Washington

32. Confidential Page 32 Confidential The Promise Capital & Land 22 Plants 54 Destination Ramps ~1,200 Load lanes ~8,400 vehicles waiting at plants $166 Million in inventory Each Plant to One Mixing Center ~22 Load lanes ~154 vehicles waiting at plants ~$3 Million in inventory

33. Confidential Page 33 Confidential The Price Inventory at the cross dock Handling at the cross dock Capital costs of the cross dock Added distance traveled

34. Confidential Page 34 Confidential Making the Trade-offs  Measuring Inventory In the rail network At the plants and Cross Docks  Load-driven system Railcars depart when full  Relationship between Network Design and Inventory

35. Confidential Page 35 Confidential Inventory at the Plants  Half a rail car full for each destination Time Laurel, Montana Orilla, Washington

36. Confidential Page 36 Confidential Inventory at the Mixing Centers  Half a rail car full for each destination Time Laurel, Montana Orilla, Washington

37. Confidential Page 37 Confidential Workload at the Mixing Centers  Unpredictable Rail car holds 5 vehicles Orilla Benicia Mira Loma Laurel Denver Orilla Benicia Mira L. Laurel Denver

38. Confidential Page 38 Confidential Workload at the Mixing Centers  Balanced: Only load cars you empty Rail car holds 5 vehicles Orilla Benicia Mira Loma Laurel Denver Orilla Benicia Mira L. Laurel Denver Orilla

39. Confidential Page 39 Confidential Effect on Inventory  Inventory at Mixing Center slowly grows to just over (ramps -1)(capacity -1) and remains there  Roughly twice the inventory of before  Still depends on the number of ramps the cross dock serves

40. Confidential Page 40 Confidential Consolidation for Speed  Unit Trains of 15-20 rail cars don’t stop at mixing yards  Trade moving inventory for stationary inventory

41. Confidential Page 41 Confidential Model  Paths  Route from Plant to Ramp  Mode used on each edge  Demand[ramp, plant]  Combined demand at ramp for all products from the plant  Variables:  PathFlow[path]: u Volume from the plant to the ramp on the path  UseLane[fromloc, toloc, mode] binary u Did we use the mode between two locations

42. Confidential Page 42 Confidential Model  Objective  Minimize the number of vehicles in the pipeline  Moving Component (Transit times)  Waiting Component (Mode Size)  Minimize PipelineInventory:  sum{path in Paths} (Total Transit Time)*PathFlow[path];  sum{(f,t,m)} (Size[m]/2)*UseLane[f,t,m]

43. Confidential Page 43 Confidential Model  Satisfy Demand  The sum of flows on all paths between a plant and a ramp must meet demand  s.t. SatisfyDemand[p in PLANTS, r in RAMPS}:  sum{path in PATHS: Plant[path]=p and Ramp[path] = r} PathFlow[path] >= Demand[p,r];

44. Confidential Page 44 Confidential Model  Define UseLane  For each pair of locations and mode between them write a constraint for each plant and ramp  s.t. DefineUseLane[p in PLANTS, r in RAMPS, (f,t,m) in EDGES}:  sum{path in PATHS: Plant[path]=p and Ramp[path] = r and (f,t,m) in PATHEDGES[path]} PathFlow[path] <= Demand[p,r]*UseLane[f,t,m];

45. Confidential Page 45 Confidential Model  Large Model  Lots of Variables: Many Paths  Lots of Constraints: DefineUseLane  The LP relaxation is nearly always integral  Use Column Generation

46. Confidential Page 46 Confidential Reduced plant destinations New Rail Lanes

47. Confidential Page 47 Confidential Mixing Centers Destination Ramps Union PacificCSXTFEC BNSFCanadian PacificCar Haul to Ramp Norfolk SouthernCanadian National Edison Norfolk Atlanta Kentucky Ohio St Louis Canada St Paul Michigan Chicago Kansas City Final Outbound Rail Network with Carriers

48. results to date

49. Confidential Page 49 Confidential Results l Cut vehicle transit time by 26% or 4 days l $1 billion savings in vehicle inventory l $125 million savings in inventory carrying costs l Avoid bottlenecks l Reduce assets in supply chain l Improved inventory turns at dealer

50. Confidential Page 50 Confidential Benefits l Ford l Dealers l Rail Carriers l Auto Haulers

51. Confidential Page 51 Confidential Benefits - Ford l On-time delivery l Competitive edge l Cost control

52. Confidential Page 52 Confidential Benefits - Dealers l Reduced inventories l Increased customer satisfaction

53. Confidential Page 53 Confidential Benefits - Rail Carriers l Improved equipment utilization (reduced capital expenditures) l Visibility and planning capabilities l Synergies with existing UPS traffic l Increased cooperation

54. Confidential Page 54 Confidential Benefits - Auto Haulers l Expanded dealer delivery hours l Visibility and planning capability l Improved asset utilization l Increased cooperation

55. Confidential Page 55 Confidential Other Opportunities Where to go from here?