Example: barber

Supply Chain Management: Logistics Network Design

Supply Chain management : Logistics NetworkDesignDonglei of Business Administration, University of New Brunswick, NB Canada FrederictonE3B 9Y2 Donglei Du (UNB)SCM1 / 67 Table of contents I1 IntroductionThe Logistics NetworkMajor Steps in Network Design2A Facility Location Problem3 Optimial Distribution and outsourcing4 Maximum flow problem5 Network synthesis problemOffline versionOnline versionDonglei Du (UNB)SCM2 / 67 Section 1 IntroductionDonglei Du (UNB)SCM3 / 67 Subsection 1 The Logistics NetworkDonglei Du (UNB)SCM4 / 67 The Logistics Network IThe objective of this chapter is to present some issues involvedin the Design and configuration of the Logistics these are strategic decisions because they have along-standing effect on the Logistics Network consists of:Facilities:Plants/Vendors P

The Logistics NetworkI The objective of this chapter is to present some issues involved in the design and con guration of the logistics network. Obviously these are strategic decisions because they have a long-standing e ect on the rm. The Logistics Network consists of: Facilities: Plants/Vendors Ports Warehouse Retailers/Distribution Centers ...

Tags:

  Network, Design, Management, Supply, Chain, Supply chain management, Logistics, Warehouse, Logistics network design

Information

Domain:

Source:

Link to this page:

Please notify us if you found a problem with this document:

Other abuse

Transcription of Supply Chain Management: Logistics Network Design

1 Supply Chain management : Logistics NetworkDesignDonglei of Business Administration, University of New Brunswick, NB Canada FrederictonE3B 9Y2 Donglei Du (UNB)SCM1 / 67 Table of contents I1 IntroductionThe Logistics NetworkMajor Steps in Network Design2A Facility Location Problem3 Optimial Distribution and outsourcing4 Maximum flow problem5 Network synthesis problemOffline versionOnline versionDonglei Du (UNB)SCM2 / 67 Section 1 IntroductionDonglei Du (UNB)SCM3 / 67 Subsection 1 The Logistics NetworkDonglei Du (UNB)SCM4 / 67 The Logistics Network IThe objective of this chapter is to present some issues involvedin the Design and configuration of the Logistics these are strategic decisions because they have along-standing effect on the Logistics Network consists of:Facilities.

2 Plants/Vendors PortsWarehouseRetailers/Distribution CentersCustomersRaw materials and finished products that flow between Du (UNB)SCM5 / 67 The Logistics Network IIplantswarehousesretailerscustomersInve ntory &warehousing costsTransportationcostTransportationcos tTransportationcostTypical Logistics ConfigurationInventory &warehousing costsProduction/purchase costsDonglei Du (UNB)SCM6 / 67 Key Strategic Decisions in the Logistics Network IAssuming that plants and retailer locations are fixed, weconcentrate on the following strategic decisions in terms the optimal number, location, and size of warehousesDetermine optimal sourcing strategyWhich plant/vendor should produce which productDetermine best distribution channelsWhich warehouses should service which retailersThe objective is to Design or reconfigure the Logistics Network soas to minimize annual system-wide costs.

3 IncludingProduction/ purchasing costsInventory carrying costs, and facility costs (handling and fixedcosts)Transportation costsDonglei Du (UNB)SCM7 / 67 Key Strategic Decisions in the Logistics Network IIThat is, we would like to find a minimal-annual-costconfiguration of the distribution Network that satisfies productdemands at specified customer service Du (UNB)SCM8 / 67 The trade-off in this problem IIncreasing the number of warehouses yieldsAn improvement in service level due to the reduction in averagetravel time to the increase in inventory costs due to increased safety stocksrequired to protect each warehouse against uncertainties incustomer increase in overhead and setup costsA reduction in outbound transportation costs: transportationcosts from the warehouse to the increase in inbound transportation costs.

4 Transportationcosts from the suppliers and/or manufacturers to the essence, the firm must balance the costs of opening newwarehouses with the advantages of being close to the Du (UNB)SCM9 / 67 The trade-off in this problem IIThus warehouse location decisions are crucial determinants ofwhether the Supply Chain is an efficient channel for thedistribution of products.$-$10$20$30$40$50$60$70$80$9002 46810 Number of WarehousesCost (millions $)Total CostTransportation CostFixed CostInventory CostOptimal Number of WarehousesDonglei Du (UNB)SCM10 / 67 Subsection 2 Major Steps in Network DesignDonglei Du (UNB)SCM11 / 67 Major Steps in Network DesignStep 1.

5 Data CollectionStep 2. Data AggregationStep 3. Data Validation and ModelStep 4. OptimizationDonglei Du (UNB)SCM12 / 67 Step 1. Data Collection IA typical Network configuration problem involves large amount ofdata, including information on1 Location of customers, stocking points and sources locationtheory2A listing of all products3 Demand for each product by customer location forecasttechnique4 Transportation rates by mode information system, like ratingengine5 Mileage estimation GIS6 Warehousing costs (handling and fixed) inventory management7 Service level requirement probabilistic technique8 Shipment sizes by productDonglei Du (UNB)SCM13 / 67 Step 1.

6 Data Collection II9 Order patterns by frequency, size, season, content10 Order processing costs11 Customer service goalsDonglei Du (UNB)SCM14 / 67 Transportation Cost ITransportation costs = Transportation rate Distance1. Transportation rate: the cost per mile per SKU. An importantcharacteristic of a class of rates for truck, rail, UPS andother trucking companies is that therates are quitelinear with the distance but not with are two kinds of transportation rates:Internal fleet (company-owned): It can be easilycalculated from information like annual costs pertruck, annual mileage per truck, annual amountdelivered, and truck s effective Du (UNB)SCM15 / 67 Transportation Cost IIExternal fleet (third-part): More complexcalculation is needed.

7 There are rating enginesavailable, such as the SMC3 RateWare Mileage estimation: Once we know the transportation rates, whichusually depends on the distance, we need to estimatethe mileage between any two locations. Depending onyour situation, you may wantExact estimation: this usually can be obtainedusing GIS system, but the drawback is cost andspeed you may need to install GIS receiver andslow down the operation of a Decision-SupportSystem (See Chapter 12 for more information)Donglei Du (UNB)SCM16 / 67 Transportation Cost IIIA pproximate estimation: For most of theapplications, this will be Du (UNB)SCM17 / 67 Warehousing costs (handling, and fixed costs) IHandling cost.

8 Proportional to the amount of material the flowsthrough the warehouseFixed Cost: All costs that are not proportional to the amount ofmaterial the flows through the warehouse . It is typicallyproportional towarehouse space size (or warehouse capacity)but in a nonlinear ,20000,40000,60000,80000,100000,800000,1 200000,150 warehouse capacitycostDonglei Du (UNB)SCM18 / 67 Warehousing costs (handling, and fixed costs) II1So we need to estimate warehouse capacity. Obviously thecapacity is proportional to the peak inventory, not the averageinventory or annual introduce the concept of inventory turnover ratio, given byinventory turnover ratio =annual flowaverage inventory level3 The warehouse capacity is given byWarehouse Capacity= 3 (2 average inventory level)= 6 annual flowinventory turnover ratioDonglei Du (UNB)SCM19 / 67 Step 2.

9 Data AggregationAggregate and clean the data because1the data collected in Step 1 is usually overwhelming,2the cost of obtaining and processing the real data is huge,3the form in which data is available must be streamlined,4the size of the resulting location model is huge, and5the accuracy of forecast demand is course, data aggregation only approximates the real data, sothe impact on model s effectiveness must be Du (UNB)SCM20 / 67 The impact of aggregate demand: 1+ 2 21+ 22 IConsider the following example with two customers: Please do ityourself in historical demands for customers 1 and 2 in the followingtable:Year199219931994199519961 9971998 Customer 122346285491956725457319862189719854 Customer 217835217651987524346228761465324987 Total 240181503143944249803548623655044841 Donglei Du (UNB)SCM21 / 67 The impact of aggregate demand: 1+ 2 21+ 22 IIHere is the summary of the historical data, we can see theaverage demand for the aggregated customer is the sum of thetwo averages.

10 However, the variability, measured by standarddeviation and coefficient of variation is deviationcoefficient of variationCustomer Du (UNB)SCM22 / 67A heuristic to aggregate data ICustomer-based Clustering: Customers located in closeproximity are aggregated using a grid Network or clusteringtechniques. All customers within a single cell or a single clusterare replaced by a single customer located at the centroid of thecell or cluster. We refer to a cell or a cluster as a customer type-based clustering: Place all SKU s into asource-group.


Related search queries