The Network Planning Process Two stonecutters were working

The Network Planning Process Two stonecutters were working on the reconstruction of St. Paul's in London when Sir Christopher Wren asked each what he was doing. The first replied, "I am cutting stone. " The second answered, "I am building a cathedral. " Christopher Wren Chapter 14 CR (2004) Prentice Hall, Inc. 1

CONTROLLING Customer service goals • The product • Logistics service • Ord. proc. & info. sys. Transport Strategy • Transport fundamentals • Transport decisions PLANNING Inventory Strategy • Forecasting • Inventory decisions • Purchasing and supply scheduling decisions • Storage fundamentals • Storage decisions ORGANIZING Network Planning in Location Strategy • Location decisions • The network planning process CR (2004) Prentice Hall, Inc. 2

The Planning Problem ·Configure the facilities of the supply chain from source points to customers ·Consider all major logistical costs, namely transportation, inventory, and facility ·Consider practical restrictions, such as capacity and customer service ·Position the analysis toward top management, strategic concerns Data for Planning ·Typical data items ·Common data sources ·Converting data to useful planning information CR (2004) Prentice Hall, Inc. 3

A Generalized Product Flow Network Sources: plants vendors ports Regional warehouses, stocking points Field warehouses, stocking points Customers, demand centers Supply Demand Supply Inventory & warehousing costs Production/ purchase costs CR (2004) Prentice Hall, Inc. Transportation costs Inventory & warehousing costs Transportation costs 14 -4 4

The Supply Channel Transportation Warehousing Transportation Customers Information flows Factory These facilities need to be positioned and sized Transportation Vendors/plants/ports Warehousing CR (2004) Prentice Hall, Inc. Transportation 14 -5 5

Geocoding Data • Linear grid • Latitude-Longitude • Other coding schemes Linear grid CR (2004) Prentice Hall, Inc. 6

Estimating Transport Costs For private trucking for accounting records for fleet Example Data Category Fact Weekly Cost Weekly mileage 2700 miles Weekly hours on duty 66 hrs. /wk. Trips per week 3 trips/wk. Driver wages $ 12. 00/hr. $792. 00 Benefits 18. 75% of wages 148. 50 Fuel cost @ 10 mpg $ 1. 10/gal 297. 00 Truck depreciation $316. 50/wk. 316. 50 Maintenance $ 45. 00/wk. 45. 00 Insurance $ 51. 00/wk. 51. 00 Tolls, food, and lodging $ 97. 50/trip 292. 50 Contingency $ 30. 00/trip 90. 00 Total $2, 032. 50 Cost per mile is $2, 032. 50/2, 700 = $0. 75 CR (2004) Prentice Hall, Inc. 7

Estimating Transport Rates For less than truckload movements CR (2004) Prentice Hall, Inc. 8

Example Shipment Profiles CR (2004) Prentice Hall, Inc. 9

Mileage Estimating By linear coordinates where DA-B = distance between points XA, YA = coordinates for point A XB, YB = coordinates for point B K = scale factor to convert the coordinate measure to a distance measure CR (2004) Prentice Hall, Inc. 10

Mileage Estimating By latitude-longitude coordinates (Great circle distance) DA-B = 3959{arccos[sin(LATA) sin(LATB) + cos(LATA) cos(LATB) cos|LONGB - LONGA|]} where DA-B = great circle distance between points A and B (statute miles) LATA= latitude of point A (radians) Corrects LONGA = longitude of point A (radians) for earth’s curvature LATB = latitude of point B (radians) LONGB= longitude of point B (radians) CR (2004) Prentice Hall, Inc. 11

Mileage Estimating Example The coordinates for Madrid, Spain are LONGA = 3. 41 W, LATA = 40. 24 N and for Milan the coordinates are LONGB = 9. 12 E, LATB = 45. 28 N. Divide degrees by 57. 3 to convert to radians. Hence, in radians LONGA = 0. 0595, LATA = 0. 7023 and LONGB = -0. 1592, LATB = 0. 7902. The distance is? Solution DA-B = 3959{arccos[sin(0. 7023) sin(0. 7902) + cos(0. 7023) cos(0. 7902) cos|-0. 1592 0. 0595|]} = 724 miles, or 1. 61 724 = 1166 kilometers Arccos, sin, and cos are found in trigonometric tables Note Multiply straight-line distance by a circuity factor to convert to approximate actual road distance. CR (2004) Prentice Hall, Inc. 12

Tools for Analysis ·Ad hoc techniques Chart/Compass/Ruler methods Spreadsheets ·Simulation models Many general-purpose models are available ·Heuristic models Used in conjunction with optimization models ·Optimization models Integer programming is popular, but usually not used alone ·AI/Expert system models No known models for this problem class CR (2004) Prentice Hall, Inc. 13

Conducting the Network Analysis ·Key questions How many facilities should there be? Where should they be located? What size should they be? ·Auditing customer service levels ·Benchmarking the current design ·Improving on the benchmark ·Seeking good designs Intuitive choices COG searching Practical considerations ·Maximum opportunity design ·Implementable design CR (2004) Prentice Hall, Inc. 14

Channel Design ·Key questions Best transport service to use? Best inventory levels to maintain? ·Use channel simulators Network and Channel Design Planning can be a hierarchical problem CR (2004) Prentice Hall, Inc. 15

Channel Simulators What they do • Replicate over time the flow of product through a supply channel on an order-by-order basis • Answer time-related questions of an operating nature, e. g. , inventory levels The Simulated Channel of SCSIM CR (2004) Prentice Hall, Inc. 16

Illustrating the “Bull Whip” Effect with a Channel Simulator Demand units 200 150 Warehouse Distributor Factory 100 Retailer 50 0 CR (2004) Prentice Hall, Inc. 5 Simulated months 10 15 17

Annually Quarterly/ monthly Monthly/ weekly Logistics Activity Network design Aggregate planning and allocation Flow planning and master production scheduling Daily Transaction processing Daily Short-term scheduling Data Base Current plans Current status Past performance Master files Hierarchical Network Supply Chain Planning Frequency 18

Closely matched No capacity restrictions Design Alternatives At current capacities a Costs are totals for three product groups. Plant capacity restrictions are at current levels but with no service restrictions. The result is direct shipments from plants. c No plant capacity or customer service restrictions. The result is direct shipments from plants. d Current plant capacities are in effect and the desired service level is set at 500 miles. e Essentially no investment in plant or warehousing is required to realize these savings. b No service restriction CR (2004) Prentice Hall, Inc. 19 14 -19

Network Design Curve High Alternative warehouse numbers, locations, and sizes Direction of improved service, no change in cost Total cost Low High Customer service Current or suboptimal network design Direction of decreased cost, no change in service Low 20 14 -20

Network Design Case Study ·Background on Aqua-Chem ·Designing the study ·Collecting and manipulating data ·Conducting the analysis ·Implementing the findings CR (2004) Prentice Hall, Inc. 21

Study Plan Define approach to product Define approach to demand geography Define current system Start Define objectives Organize project team Select project approach Define approach to material flow End phase 1 Define approach to cost elements CR (2004) Prentice Hall, Inc. 22

Typical Data Items • Products in the product line • Locations of customers, warehouses, and plants • Demand for each customer by location and product • Transportation rates • Transit times • Warehousing costs (fixed, storage, and handling) • Purchase/production costs • Shipment sizes by product • Inventory levels by location, by product, and control policies • Shipment profiles • Order processing cost • Capital cost • Customer service goals • Available and potential facilities with capacities • Distribution patterns for current product flows CR (2004) Prentice Hall, Inc. 23

Small orders Store at warehouse Freight to warehouse + Storage cost + Handling cost + Order processing cost + Inventory cost + Personal property tax + Freight to customer CR (2004) Prentice Hall, Inc. Large orders Store at plant Ship direct Order processing cost Storage cost + + Freight to customer Handling cost + Order processing cost + Inventory cost + Personal property tax + Freight to customer Customer Alternative Distribution Channels for Aqua-Chem Purchase cost of raw materials + Freight to producing plant + Production labor cost + Variable overhead 24

Aqua-Chem Network Design Results CR (2004) Prentice Hall, Inc. 25 14 -25