Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Ford Motor Company’s Finished Vehicle Distribution System April 2001 Ellen Ewing Project Director UPS Logistics Dr. John V...
Agenda <ul><ul><li>Introduction </li></ul></ul><ul><ul><li>1999 Environment </li></ul></ul><ul><ul><li>Solution Approach  ...
Introduction
Objectives/Motivation <ul><ul><li>Importance of Pipeline Inventory </li></ul></ul><ul><ul><li>Pipeline Inventory and Netwo...
Competitive Necessity <ul><li>The new BMW Sales and Production System </li></ul>
<ul><ul><li>Financial Incentives: Capital Utilization </li></ul></ul><ul><ul><ul><li>In 1996 </li></ul></ul></ul><ul><ul><...
The Need for Speed <ul><li>Demand for land </li></ul><ul><li>22 Plants </li></ul><ul><li>54 Destination Ramps </li></ul><u...
The Need for Speed <ul><li>Other Incentives </li></ul><ul><ul><li>Damage </li></ul></ul><ul><ul><li>Flexibility </li></ul>...
Before 1996
 
The Price <ul><li>Inventory at the cross dock </li></ul><ul><li>Added distance traveled </li></ul><ul><li>Handling at the ...
Mixing Centers
1999  environment
1999 Vehicle Network Delivery Conditions <ul><ul><li>Record production levels </li></ul></ul><ul><ul><li>Demand shift from...
High 1999 Level Statistics <ul><ul><li>Assembly plants 22 </li></ul></ul><ul><ul><li>Mixing centers   5 </li></ul></ul><ul...
Ford Distribution Network Mixing Center Origin Plant Groupings Destination Ramp Planned Ramp Closure Edison Norfolk Atlant...
Old Delivery Design <ul><ul><li>Push Network </li></ul></ul><ul><ul><li>Vendor sub systems optimized for individual segmen...
solution approach
Ford Goals <ul><li>Speed </li></ul><ul><ul><li>1999: Average 15 days transit time  </li></ul></ul><ul><ul><li>Goal: Maximu...
network  design
Design Process <ul><li>Truck vs Rail delivery </li></ul><ul><li>Allocate Dealers (FIPS)  </li></ul><ul><li>to Ramps </li><...
Ford Locations Plant Mixing Center Origin Ramp Dest. Ramp MC Ramp
An Allocation Model
Single-Sourcing Allocation <ul><li>Var Assign{FIPS, RAMPS} binary; </li></ul><ul><li>Minimize TotalCost: </li></ul><ul><li...
Old Ramp Allocation Southern US Dealers sourced by   multiple ramps
New Ramp Allocation Southern US Dealers sourced by single ramps
New Allocation of Dealers to Ramps Mainland US
Flows through the Rail Network <ul><li>Objective is NOT Freight cost! </li></ul>
The Objective IS <ul><li>Speed </li></ul><ul><li>Capital </li></ul><ul><li>Land </li></ul>
The Promise <ul><li>Speed </li></ul><ul><li>Unit trains bypass hump yards </li></ul>
The Promise <ul><li>Capital & Land </li></ul>Time Laurel, Montana Orilla, Washington
The Promise <ul><li>Capital & Land </li></ul><ul><li>22 Plants </li></ul><ul><li>54 Destination Ramps </li></ul><ul><li>~1...
The Price <ul><li>Inventory at the cross dock </li></ul><ul><li>Handling at the cross dock </li></ul><ul><li>Capital costs...
Making the Trade-offs <ul><li>Measuring Inventory </li></ul><ul><li>In the rail network  </li></ul><ul><li>At the plants a...
Inventory at the Plants <ul><li>Half a rail car full for each destination </li></ul>Time Laurel, Montana Orilla, Washington
Inventory at the Mixing Centers <ul><li>Half a rail car full for each destination </li></ul>Time Laurel, Montana Orilla, W...
Workload at the Mixing Centers <ul><li>Unpredictable </li></ul>Rail car holds 5 vehicles Orilla Benicia Mira Loma Laurel D...
Workload at the Mixing Centers <ul><li>Balanced: Only load cars you empty </li></ul>Rail car holds 5 vehicles Orilla Benic...
Effect on Inventory <ul><li>Inventory at Mixing Center slowly grows  </li></ul><ul><li>to just over (ramps -1)(capacity -1...
Consolidation for Speed <ul><li>Unit Trains of 15-20 rail cars don’t stop at  </li></ul><ul><li>mixing yards </li></ul><ul...
Model <ul><li>Paths </li></ul><ul><ul><li>Route from Plant to Ramp </li></ul></ul><ul><ul><li>Mode used on each edge </li>...
Model <ul><li>Objective </li></ul><ul><ul><li>Minimize the number of vehicles in the pipeline </li></ul></ul><ul><ul><ul><...
Model <ul><li>Satisfy Demand </li></ul><ul><ul><li>The sum of flows on all paths between a plant and a ramp must meet dema...
Model <ul><li>Define UseLane </li></ul><ul><ul><li>For each pair of locations and mode between them write a constraint for...
Model <ul><li>Large Model  </li></ul><ul><ul><li>Lots of Variables: Many Paths </li></ul></ul><ul><ul><li>Lots of Constrai...
New Rail Lanes Reduced plant destinations
Final Outbound Rail Network with Carriers Mixing Centers Destination Ramps Union Pacific CSXT FEC BNSF Canadian Pacific Ca...
results to date
Results <ul><ul><li>Cut vehicle transit time by 26% or 4 days </li></ul></ul><ul><ul><li>$1 billion savings in vehicle inv...
Benefits  <ul><li>Ford  </li></ul><ul><li>Dealers  </li></ul><ul><li>Rail Carriers  </li></ul><ul><li>Auto Haulers   </li>...
Benefits - Ford <ul><li>On-time delivery </li></ul><ul><li>Competitive edge </li></ul><ul><li>Cost control </li></ul>
Benefits - Dealers <ul><li>Reduced inventories </li></ul><ul><li>Increased customer satisfaction </li></ul>
Benefits - Rail Carriers <ul><li>Improved equipment utilization (reduced capital  </li></ul><ul><li>expenditures) </li></u...
Benefits - Auto Haulers <ul><li>Expanded dealer delivery hours </li></ul><ul><li>Visibility and  planning capability </li>...
Other Opportunities <ul><li>Where to go from here? </li></ul>
 
Upcoming SlideShare
Loading in …5
×

Pipeline Inventory

0 views

Published on

  • Be the first to comment

  • Be the first to like this

Pipeline Inventory

  1. 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. 2. Agenda <ul><ul><li>Introduction </li></ul></ul><ul><ul><li>1999 Environment </li></ul></ul><ul><ul><li>Solution Approach </li></ul></ul><ul><ul><li>Network Design </li></ul></ul><ul><ul><li>Implement New Strategy </li></ul></ul><ul><ul><li>Results to Date </li></ul></ul><ul><ul><li>Summary </li></ul></ul>
  3. 3. Introduction
  4. 4. Objectives/Motivation <ul><ul><li>Importance of Pipeline Inventory </li></ul></ul><ul><ul><li>Pipeline Inventory and Network Design </li></ul></ul><ul><ul><li>Role of modeling </li></ul></ul><ul><ul><li>Information in variables </li></ul></ul><ul><ul><li>Stronger formulation </li></ul></ul><ul><ul><li>Financial impact </li></ul></ul>
  5. 5. Competitive Necessity <ul><li>The new BMW Sales and Production System </li></ul>
  6. 6. <ul><ul><li>Financial Incentives: Capital Utilization </li></ul></ul><ul><ul><ul><li>In 1996 </li></ul></ul></ul><ul><ul><ul><ul><li>Ford produced 3.9 million vehicles in the US </li></ul></ul></ul></ul><ul><ul><ul><ul><li>Avg. transit time 15+ days </li></ul></ul></ul></ul><ul><ul><ul><ul><li>Avg. vehicle revenue $18,000 </li></ul></ul></ul></ul><ul><ul><ul><ul><li>Value of pipeline inventory: > $2.8 Billion </li></ul></ul></ul></ul><ul><ul><ul><ul><li>One day reduced transit time: </li></ul></ul></ul></ul><ul><ul><ul><ul><ul><li>$190 Million reduction in pipeline inv. </li></ul></ul></ul></ul></ul><ul><ul><ul><ul><ul><li>1,400 fewer railcars </li></ul></ul></ul></ul></ul>The Need for Speed
  7. 7. The Need for Speed <ul><li>Demand for land </li></ul><ul><li>22 Plants </li></ul><ul><li>54 Destination Ramps </li></ul><ul><li>~1,200 Load lanes </li></ul><ul><li>~8,400 vehicles waiting at plants </li></ul><ul><li>$166 Million in inventory </li></ul>
  8. 8. The Need for Speed <ul><li>Other Incentives </li></ul><ul><ul><li>Damage </li></ul></ul><ul><ul><li>Flexibility </li></ul></ul><ul><ul><li>Others? </li></ul></ul>
  9. 9. Before 1996
  10. 11. The Price <ul><li>Inventory at the cross dock </li></ul><ul><li>Added distance traveled </li></ul><ul><li>Handling at the cross dock </li></ul><ul><li>Capital costs of the cross dock </li></ul>
  11. 12. Mixing Centers
  12. 13. 1999 environment
  13. 14. 1999 Vehicle Network Delivery Conditions <ul><ul><li>Record production levels </li></ul></ul><ul><ul><li>Demand shift from cars to trucks </li></ul></ul><ul><ul><li>Overburdened rail infrastructure </li></ul></ul><ul><ul><li>Deteriorating rail service </li></ul></ul><ul><ul><li>Shortage of transport capacity </li></ul></ul><ul><ul><li>Mixing centers </li></ul></ul><ul><ul><li>15+ day transit time </li></ul></ul><ul><ul><li>High inventory cost </li></ul></ul><ul><ul><li>Dissatisfied customers </li></ul></ul>
  14. 15. High 1999 Level Statistics <ul><ul><li>Assembly plants 22 </li></ul></ul><ul><ul><li>Mixing centers 5 </li></ul></ul><ul><ul><li>Destination rail ramps 54 </li></ul></ul><ul><ul><li>Dealer locations 6,000 </li></ul></ul><ul><ul><li>Production volume 4.4 Mil./Year </li></ul></ul><ul><ul><li>Freight expense $1.5 Bil. </li></ul></ul><ul><ul><li>Dec. ‘99 avg. transit time 16.8 Days </li></ul></ul><ul><ul><li>Pipeline Inventory $4.1 Bil. </li></ul></ul>
  15. 16. Ford Distribution Network Mixing Center Origin Plant Groupings Destination Ramp Planned 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)
  16. 17. Old Delivery Design <ul><ul><li>Push Network </li></ul></ul><ul><ul><li>Vendor sub systems optimized for individual segments </li></ul></ul><ul><ul><li>Little to no visibility </li></ul></ul><ul><ul><li>Mixing Centers not used effectively </li></ul></ul>
  17. 18. solution approach
  18. 19. Ford Goals <ul><li>Speed </li></ul><ul><ul><li>1999: Average 15 days transit time </li></ul></ul><ul><ul><li>Goal: Maximum of 8 days transit time </li></ul></ul><ul><li>Precision </li></ul><ul><ul><li>1998/1999: 37% on time within 1 week </li></ul></ul><ul><ul><li>Goal: 95% on time within 1 day </li></ul></ul><ul><li>Visibility </li></ul><ul><ul><li>100 % Internet vehicle tracking from plant release to dealer delivery </li></ul></ul><ul><ul><li>Guide the flow of vehicles </li></ul></ul><ul><ul><li>Respond to variations </li></ul></ul><ul><ul><li>Inform customers </li></ul></ul>
  19. 20. network design
  20. 21. Design Process <ul><li>Truck vs Rail delivery </li></ul><ul><li>Allocate Dealers (FIPS) </li></ul><ul><li>to Ramps </li></ul><ul><li>Route Flows through </li></ul><ul><li>Rail Network </li></ul>
  21. 22. Ford Locations Plant Mixing Center Origin Ramp Dest. Ramp MC Ramp
  22. 23. An Allocation Model
  23. 24. Single-Sourcing Allocation <ul><li>Var Assign{FIPS, RAMPS} binary; </li></ul><ul><li>Minimize TotalCost: </li></ul><ul><li>sum{fip in FIPS,ramp in RAMPS} Cost[fip,ramp]*Assign[fip,ramp]; </li></ul><ul><li>s.t. SingleSource{fip in FIPS}: </li></ul><ul><li>sum{ramp in RAMPS}Assign[fip,ramp] = 1; </li></ul><ul><li>s.t. ObserveCapacity{ramp in RAMPS}: </li></ul><ul><li>sum{fip in FIPS} Volume[fip]*Assign[fip,ramp] </li></ul><ul><li><= Capacity[ramp]; </li></ul>
  24. 25. Old Ramp Allocation Southern US Dealers sourced by multiple ramps
  25. 26. New Ramp Allocation Southern US Dealers sourced by single ramps
  26. 27. New Allocation of Dealers to Ramps Mainland US
  27. 28. Flows through the Rail Network <ul><li>Objective is NOT Freight cost! </li></ul>
  28. 29. The Objective IS <ul><li>Speed </li></ul><ul><li>Capital </li></ul><ul><li>Land </li></ul>
  29. 30. The Promise <ul><li>Speed </li></ul><ul><li>Unit trains bypass hump yards </li></ul>
  30. 31. The Promise <ul><li>Capital & Land </li></ul>Time Laurel, Montana Orilla, Washington
  31. 32. The Promise <ul><li>Capital & Land </li></ul><ul><li>22 Plants </li></ul><ul><li>54 Destination Ramps </li></ul><ul><li>~1,200 Load lanes </li></ul><ul><li>~8,400 vehicles waiting at plants </li></ul><ul><li>$166 Million in inventory </li></ul><ul><li>Each Plant to One Mixing Center </li></ul><ul><li>~22 Load lanes </li></ul><ul><li>~154 vehicles waiting at plants </li></ul><ul><li>~$3 Million in inventory </li></ul>
  32. 33. The Price <ul><li>Inventory at the cross dock </li></ul><ul><li>Handling at the cross dock </li></ul><ul><li>Capital costs of the cross dock </li></ul><ul><li>Added distance traveled </li></ul>
  33. 34. Making the Trade-offs <ul><li>Measuring Inventory </li></ul><ul><li>In the rail network </li></ul><ul><li>At the plants and Cross Docks </li></ul><ul><li>Load-driven system </li></ul><ul><li>Railcars depart when full </li></ul><ul><li>Relationship between </li></ul><ul><li>Network Design and Inventory </li></ul>
  34. 35. Inventory at the Plants <ul><li>Half a rail car full for each destination </li></ul>Time Laurel, Montana Orilla, Washington
  35. 36. Inventory at the Mixing Centers <ul><li>Half a rail car full for each destination </li></ul>Time Laurel, Montana Orilla, Washington
  36. 37. Workload at the Mixing Centers <ul><li>Unpredictable </li></ul>Rail car holds 5 vehicles Orilla Benicia Mira Loma Laurel Denver Orilla Benicia Mira L. Laurel Denver
  37. 38. Workload at the Mixing Centers <ul><li>Balanced: Only load cars you empty </li></ul>Rail car holds 5 vehicles Orilla Benicia Mira Loma Laurel Denver Orilla Orilla Benicia Mira L. Laurel Denver
  38. 39. Effect on Inventory <ul><li>Inventory at Mixing Center slowly grows </li></ul><ul><li>to just over (ramps -1)(capacity -1) and </li></ul><ul><li>remains there </li></ul><ul><li>Roughly twice the inventory of before </li></ul><ul><li>Still depends on the number of ramps the </li></ul><ul><li>cross dock serves </li></ul>
  39. 40. Consolidation for Speed <ul><li>Unit Trains of 15-20 rail cars don’t stop at </li></ul><ul><li>mixing yards </li></ul><ul><li>Trade moving inventory for stationary </li></ul><ul><li>inventory </li></ul>
  40. 41. Model <ul><li>Paths </li></ul><ul><ul><li>Route from Plant to Ramp </li></ul></ul><ul><ul><li>Mode used on each edge </li></ul></ul><ul><li>Demand[ramp, plant] </li></ul><ul><ul><li>Combined demand at ramp for all </li></ul></ul><ul><ul><li>products from the plant </li></ul></ul><ul><li>Variables: </li></ul><ul><ul><li>PathFlow[path]: </li></ul></ul><ul><ul><ul><li>Volume from the plant to the ramp on </li></ul></ul></ul><ul><ul><ul><li>the path </li></ul></ul></ul><ul><ul><li>UseLane[fromloc, toloc, mode] binary </li></ul></ul><ul><ul><ul><li>Did we use the mode between two locations </li></ul></ul></ul>
  41. 42. Model <ul><li>Objective </li></ul><ul><ul><li>Minimize the number of vehicles in the pipeline </li></ul></ul><ul><ul><ul><li>Moving Component (Transit times) </li></ul></ul></ul><ul><ul><ul><li>Waiting Component (Mode Size) </li></ul></ul></ul><ul><ul><li>Minimize PipelineInventory: </li></ul></ul><ul><ul><ul><li>sum{path in Paths} (Total Transit Time)*PathFlow[path]; </li></ul></ul></ul><ul><ul><ul><li>sum{(f,t,m)} (Size[m]/2)*UseLane[f,t,m] </li></ul></ul></ul>
  42. 43. Model <ul><li>Satisfy Demand </li></ul><ul><ul><li>The sum of flows on all paths between a plant and a ramp must meet demand </li></ul></ul><ul><ul><li>s.t. SatisfyDemand[p in PLANTS, r in RAMPS}: </li></ul></ul><ul><ul><ul><li>sum{path in PATHS: Plant[path]=p and Ramp[path] = r} </li></ul></ul></ul><ul><ul><ul><li>PathFlow[path] >= Demand[p,r]; </li></ul></ul></ul>
  43. 44. Model <ul><li>Define UseLane </li></ul><ul><ul><li>For each pair of locations and mode between them write a constraint for each plant and ramp </li></ul></ul><ul><ul><li>s.t. DefineUseLane[p in PLANTS, r in RAMPS, </li></ul></ul><ul><ul><li>(f,t,m) in EDGES}: </li></ul></ul><ul><ul><ul><li>sum{path in PATHS: Plant[path]=p and </li></ul></ul></ul><ul><ul><ul><li>Ramp[path] = r and </li></ul></ul></ul><ul><ul><ul><li>(f,t,m) in PATHEDGES[path]} </li></ul></ul></ul><ul><ul><ul><li>PathFlow[path] <= Demand[p,r]*UseLane[f,t,m]; </li></ul></ul></ul>
  44. 45. Model <ul><li>Large Model </li></ul><ul><ul><li>Lots of Variables: Many Paths </li></ul></ul><ul><ul><li>Lots of Constraints: DefineUseLane </li></ul></ul><ul><li>The LP relaxation is nearly always integral </li></ul><ul><li>Use Column Generation </li></ul>
  45. 46. New Rail Lanes Reduced plant destinations
  46. 47. Final Outbound Rail Network with Carriers Mixing Centers Destination Ramps Union Pacific CSXT FEC BNSF Canadian Pacific Car Haul to Ramp Norfolk Southern Canadian National Edison Norfolk Atlanta Kentucky Ohio St Louis Canada St Paul Michigan Chicago Kansas City
  47. 48. results to date
  48. 49. Results <ul><ul><li>Cut vehicle transit time by 26% or 4 days </li></ul></ul><ul><ul><li>$1 billion savings in vehicle inventory </li></ul></ul><ul><ul><li>$125 million savings in inventory carrying costs </li></ul></ul><ul><ul><li>Avoid bottlenecks </li></ul></ul><ul><ul><li>Reduce assets in supply chain </li></ul></ul><ul><ul><li>Improved inventory turns at dealer </li></ul></ul>
  49. 50. Benefits <ul><li>Ford </li></ul><ul><li>Dealers </li></ul><ul><li>Rail Carriers </li></ul><ul><li>Auto Haulers </li></ul>
  50. 51. Benefits - Ford <ul><li>On-time delivery </li></ul><ul><li>Competitive edge </li></ul><ul><li>Cost control </li></ul>
  51. 52. Benefits - Dealers <ul><li>Reduced inventories </li></ul><ul><li>Increased customer satisfaction </li></ul>
  52. 53. Benefits - Rail Carriers <ul><li>Improved equipment utilization (reduced capital </li></ul><ul><li>expenditures) </li></ul><ul><li>Visibility and planning capabilities </li></ul><ul><li>Synergies with existing UPS traffic </li></ul><ul><li>Increased cooperation </li></ul>
  53. 54. Benefits - Auto Haulers <ul><li>Expanded dealer delivery hours </li></ul><ul><li>Visibility and planning capability </li></ul><ul><li>Improved asset utilization </li></ul><ul><li>Increased cooperation </li></ul>
  54. 55. Other Opportunities <ul><li>Where to go from here? </li></ul>

×