Supply Chain Management Concepts

1
Supply Chain Management Concepts

• OEM 2001
• Joe Geunes

2
Supply Chain Management and Analysis

• What is Supply Chain Management (SCM)?
• What is the difference (if any) between SCM and
  Business Logistics Management?
• Supply Chain Definition (G.C. Stevens, 1989) .
  . . a connected series of activities which is
  concerned with planning, coordinating and
  controlling materials, parts, and finished goods
  from supplier to customer. It is concerned with
  two distinct flows (material and information)
  through the organization.
• The Basic Problem Get the right amounts of the
  right products to the right markets at the right
  time in the most economical way.

3
Key Supply Chain Activities

• Customer Service Standards
• Cooperate with marketing to
• Determine customer needs and wants for logistics
  customer service
• Determine customer response to service
• Set customer service levels
• Transportation
• Mode and transport service selection
• Freight consolidation
• Carrier routing
• Vehicle scheduling
• Equipment selection
• Claims processing
• Rate auditing

4
Key Supply Chain Activities

• Inventory management
• Raw materials and finished goods stocking
  policies
• Short-term sales forecasting
• Product mix at stocking points
• Number, size, and location of stocking points
• Just-in-time, push, and pull strategies
• Information flows and order processing
• Sales order-inventory interface procedures
• Order information transmittal methods
• Ordering rules

5
Key Supply Chain Support Activities

• Warehousing
• Space determination
• Stock layout and dock design
• Warehouse configuration
• Stock placement
• Materials handling
• Equipment selection
• Equipment replacement policies
• Order-picking procedures
• Stock storage and retrieval

6
Key Supply Chain Support Activities

• Purchasing
• Supply source selection
• Purchase timing
• Purchase quantities
• Protective package design for
• Handling
• Storage
• Protection from loss and damage
• Cooperate with production/operations to
• Specify aggregate quantities
• Sequence and time production output

7
Key Supply Chain Support Activities

• Information maintenance
• Information collection, storage, and manipulation
• Data analysis
• Control procedures
• We cant forget additional factors related to
  product design for manufacture and distribution,
  i.e., the constraints that product
  characteristics place on ease of manufacture and
  distribution. Another important issue is product
  mix from a marketing standpoint, i.e., which
  products the chain will carry.

8
Logistics Strategy and Planning

• Three objectives of logistics strategy
• Cost reduction (variable costs)
• Capital reduction (investment, fixed costs)
• Service Improvement (may be at odds with the
  above two objectives).
• Major Logistics Planning Areas
• Customer service goals (customer
  requirements/costs of providing service)
• Facility location strategy (assign markets to
  plants to minimize distribution costs)
• Inventory decisions (push, pull, location of
  stocks)
• Transport strategy (modes, shipment sizes,
  routing, scheduling)

9
Considerations in Formulating Strategy

• Demand (volume, dispersion, predictability)
• Customer service requirements (customer
  expectations and competition)
• Product characteristics (density, value, risk)
• Logistics costs (based on above three factors)
• Pricing policy (does the customer price include
  delivery charge?)

10
Typical Supply Chain Strategies

• Postponement Delay product differentiation
  until as late as possible in the production
  process
• HP printers that serve different countries used
  to be produced as separate products, but now the
  same product uses an external power pack that is
  packed in the box depending on the destination
• Benetton delays dying of fabrics until after the
  sweater is produced and demand is realized).
• Postponement usually involves products with a
  highly modular architecture (e.g., Gateway and
  Dell computers).

11
Typical Strategies

• Consolidation if the market demands several
  products made by the manufacturer, consolidating
  them into one warehouse will make it more
  economical to send frequent consolidated
  shipments of full truckloads to the market.
• Mass Customization a modular product
  architecture helps enable mass customization,
  which is the ability to mass produce goods that
  can quickly and easily be customized to
  individual specifications (as in the Gateway and
  Dell computer examples).
• JIT/VMI - Just-in-time and vendor managed
  inventory strategies to smooth flow of goods and
  increase response time of suppliers.

12
Product Characteristics

• Product life cycle
• 80-20 rule
• Individual characteristics
• Weight-Bulk ratio (ration of weight to volume,
  density e.g. cotton vs. steel)
• Value-Weight ration (coal vs. jewelry)
• Substitutability (customers reaction when not in
  stock)
• Risk characteristics (perishability,
  flammability, ease of being stolen)

13
Customer Service Elements

• Pretransaction elements
• Written policies
• System flexibility
• Clarity of procedures
• Technical help
• Transaction elements
• Backorder policies
• Order cycle time (lead time)
• Product substitution
• Complexity of transaction (convenience)

14
Customer Service Elements

• Post-transaction elements
• Installation, warranty, repairs Claims,
  complaints Packaging Temporary replacement
  during repair
• Courtesy, Reliability and integrity
• Willingness to respond to customer wants and
  needs (with new or better products)
• Clarity of communications to customer
• Integrated information systems
• A monopolistic company must also adhere to these
  guidelines in case competition strikes in the
  future (e.g. ATT, cable, utility companies)

15
Customer Service Aspects of Logistics

• Order cycle time time between placing order and
  receiving product
• Order transmittal
• Order processing and assembly
• Additional stock acquisition time (if out of
  stock)
• Delivery time
• On the average it is approximately six times
  more expensive to develop a new customer than it
  is to keep a current customer. Thus, from a
  financial point of view, resources invested in
  customer service activities provide a
  substantially higher return than resources
  invested in promotion and other customer
  development activities.
• P.S. Bender, Design and Operation of Customer
  Service Systems, 1976.

16
Cost vs. Service Models
17
Transport Fundamentals

• Transport involves equipment (trucks, planes,
  trains, boats, pipeline), people (drivers,
  loaders unloaders), and decisions (routing,
  timing, quantities, equipment size, transport
  mode). In underdeveloped countries we often find
  it necessary to locate production close to both
  markets and resources, while in countries with
  developed distribution systems people can live in
  places far from production and resources.
• When deciding the transport mode for a given
  product there are several things to consider
• Mode price
• Transit time and variability (reliability)
• Potential for loss or damage

18
Single-mode Service Choices and Issues

• Rail (long distance, heavy goods, slow mover)
  Carload (CL) vs. less-than-carload (LCL per
  hundredweight cwt.)
• avg. length of haul 720 miles
• avg. speed 22 mph
• Larger cars can carry around 83 tons
• Truck (Smaller goods than rail, medium time
  duration)
• avg. 646 miles for truckload (TL), 274 miles for
  less-than-truckload (LTL)
• More than ½ shipments are less than 10,000 lbs.
• Trucks can go door-to-door as opposed to planes
  and trains
• Can hold 30-50,000 lbs. depending on the product
  density
• avg. 35-45 mph

19
Single-mode Service Choices and Issues

• Air (Smallest size goods, quick transport)
• avg. 545-585 mph
• avg. distance of 1,300 miles
• Low variability in lead time
• Requires transport to and from airport
• Water (Extremely slow, large goods, international
  trade)
• avg. speed on Mississippi 5 9 mph
• avg. distance 500 miles on rivers, 550 miles on
  Great Lakes, 1775 miles coast lines
• Up to 40,000 tons
• Pipeline (limited product line, liquids, gases)
• 3 4 mph (89,000 gallons per hour in a 1ft
  diameter pipe)
• Highly reliable
• Low product losses

20
Transport Cost Characteristics

• Fixed costs
• Terminal facilities
• Transport equipment
• Carrier administration
• Roadway acquisition and maintenance
• Variable costs
• Fuel
• Labor
• Equipment maintenance
• Handling, pickup, and delivery

21
Transport Cost Characteristics

• Rail
• High fixed costs, low variable costs
• High volumes result in lower per unit (variable)
  costs
• Highway
• Lower fixed costs (dont need to own or maintain
  roads)
• Higher unit costs than rail due to lower capacity
  per truck
• Terminal expenses and line-haul expenses
• Water
• High terminal (port) costs and high equipment
  costs (both fixed)
• Very low unit costs
• Air
• Substantial fixed costs
• Variable costs depend highly on distance traveled
• Pipeline
• Highest proportion of fixed cost of any mode due
  to pipeline ownership and maintenance and
  extremely low variable costs

22
Transportation Rate Structures

• Volume-based rates (based on weight)
• Distance-based rates
• Typically some combination of both of the above
• Important that rates are consistent and
  relatively simple
• Simplest rate US Mail first class letter rate
• Typical rate charts based on distance and weight
• Freight class also very important the class
  of an item depends on its density and bulkiness

23
Vehicle Routing

• Separate single origin and destination
• Once we have selected a transport mode and have
  goods that need to go from point A to point B, we
  must decide how to route a vehicle (or vehicles)
  from point A to point B.
• Given a map of all of our route choices between A
  and B we can create a network representing these
  choices The problem then reduces to the problem
  of finding the shortest path in the network from
  point A to B.
• This is a well solved problem that can use
  Dijkstras Algorithm for quick solution of small
  to medium (several thousand nodes) sized
  problems.

24
Multiple Origin and Destination Points

• Suppose we have multiple sources and multiple
  destinations, that each destination requires some
  integer number of truckloads, and that none of
  the sources have capacity restrictions. In this
  case we can simply apply the transportation
  method of linear programming to determine the
  assignment of sources to destinations.

25
Transportation Problem Formulation

• Minimize
• Subject to

26
Coincident Origin and Destination The TSP

• If a vehicle must deliver to more than two
  customers, we must decide the order in which we
  will visit those customers so as to minimize the
  total cost of making the delivery.
• We first suppose that any time that we make a
  delivery to customers we are able to make use of
  only a single vehicle, i.e., that vehicle
  capacity of our only truck is never an issue.
• In this case, we need to dispatch a single
  vehicle from our depot to n - 1 customers, with
  the vehicle returning to the depot following its
  final delivery.
• This is the well-known Traveling Salesman Problem
  (TSP). The TSP has been well studied and solved
  for problem instances involving thousands of
  nodes. We can formulate the TSP as follows

27
TSP Formulation

• Minimize
• Subject to

28
TSP Formulation

• In the TSP formulation if we remove the third
  constraint set we have the simple assignment
  problem, which can be easily solved.
• The addition of the third constraint set,
  commonly called subtour elimination constraints,
  makes this a very difficult problem to solve.

29
Questions about the TSP

• Given a problem with n nodes, how many distinct
  feasible tours exist?
• How many arcs will the network have?
• How many xij variables will we have?
• How could we quantify the number of subtour
  elimination constraints?
• The complexity of the TSP has led to several
  heuristic or approximate methods for finding good
  feasible solutions. The simplest solution we
  might think of is that of the nearest neighbor.

30
6 City TSP Network
Illustration of subtours
31
TSP Heuristics

• A second heuristic, known as the sweep heuristic,
  will perform much better in the worst case then
  the nearest neighbor.
• The sweep heuristic basically attempts to make an
  outer loop around the nodes.
• Draw a straight line emanating from the depot
  with a length r which is at least as great in
  length as the maximum straight-line distance from
  the depot to any customer (the direction of the
  line is not important).
• Visualize the line as sweeping either clockwise
  or counter-clockwise through a circle of radius
  r. Each time the radius line intersects a
  customer location we make that customer the next
  customer on the route.

32
Single Depot, Multiple Destinations, Vehicle
Capacities

• When the depot contains many vehicles and vehicle
  capacity constraints come into play, the problem
  becomes even more complex.
• If each customer has enough demand to receive a
  full truckload the problem is easy and we simply
  use the shortest path to get the single truck to
  each customer. Otherwise, we must decide which
  customers will receive deliveries from the same
  truck, and then we must decide how to route the
  trucks to the customers on the route.
• We will look at a mixed-integer programming
  formulation of the Vehicle Routing Problem (VRP).

33
Illustration of VRP
34
The Vehicle Routing Problem (VRP)

• The Vehicle routing problem (VRP) generalizes the
  TSP since we have a set of K capacity constrained
  (homogeneous) vehicles at a depot, each of which
  must visit a subset of the n - 1 customers
  exactly once and return to the depot.
• No two vehicles may visit the same customer.
  This means that each vehicle must complete a
  Hamiltonian tour (a Hamiltonian tour is a
  feasible TSP solution).
• The objective is to determine the minimum travel
  cost required to serve each customer. Let A
  denote the set of pairs of cities, and let k
  index trucks, each with capacity u. Assume that
  customer i has demand equal to di.

35
VRP Heuristics

• Given the difficulties in solving the TSP
  problem, we cannot expect to have great success
  solving VRP problems without some sort of
  heuristic approaches. We can use several guiding
  principles in developing these heuristics. (Note
  that the above formulation does not consider
  additional practical restrictions such as limits
  on driver time, time window delivery
  restrictions, or return of goods from customers
  to the depot.)

36
VRP Heuristic Principles

• 1. Try to assign customers in close proximity to
  the same truck.
• 2. Assign customers in close proximity (not on
  the same truck) to the same delivery day (to
  better manage capacity usage).
• 3. Build routes beginning with the farthest
  delivery and cluster around this delivery first.
• 4. Routes should form a teardrop pattern
  (similar to sweep heuristic for TSP).
• 5. Allocate largest vehicles to routes before
  small vehicles.
• 6. Plan pickups during deliveries, not after all
  deliveries have been made.
• 7. Outliers are candidates for alternate means
  of transport.
• 8. Avoid time windows if possible.

37
VRP Sweep Heuristic

• Note that the sweep method, when applied to the
  VRP, will have a slightly different
  interpretation. That is, we can only add a
  delivery location to a route as long as it does
  not exceed the vehicle capacity. So we can only
  continue to assign deliveries to a route as long
  as the vehicle capacity is not exceeded. Then we
  need to start assigning deliveries to a new truck.

38
Sweep Heuristic
Start Sweep
39
Facility Location Decisions

• Classifying location decisions
• Driving force (critical factor - traffic, labor
  rates, emergency facilities, obnoxious
  facilities)
• Number of facilities
• Discrete vs. continuous choices
• Data aggregation
• Time Horizon

40
Facility Location

• Rent Curve - The rent of land is a decreasing
  function of the distance to the market
• Weight gaining vs. weight losing industries
• Weight losing should locate close to raw
  materials
• Weight gaining should locate close to market
• Tapered (concave) transportation costs
• The derivative of total transportation cost is
  non-increasing with the distance to the market
  (holds for inbound and outbound costs)
• Optimal solution will always locate either at raw
  materials or at market (extreme point solution)

41
Single Facility Location Model

• This model assumes a known set, I, of source and
  demand points, each with known demand volumes,
  Vi, and transportation rates, Ri.
• The objective is to locate the facility at the
  point that minimizes total transportation cost,
  TC
• Let di denote the distance from the facility to
  demand point i.
• Min
• subject to
• The decision variables are the coordinates of
  the facility
• Xi, Yi denote the coordinates of demand point i.

42
Single Facility Location Model

• Differentiating TC w.r.t. and setting the result
  equal to zero gives the center of gravity

43
Single Facility Location Model

• This continuous problem is often called the Weber
  problem
• These problems are restrictive because they
  assume continuity of location and straight-line
  distances
• Also, only variable distance related costs are
  considered

44
General Facility Location Model

• The general facility location problem considers
  the simultaneous location of a number of
  facilities
• Notation
• I - Set of customers, indexed by i.
• J - Set of facilities, indexed by j.
• di - demand of customer i.
• cij - cost of transporting a unit from facility j
  to customer i.
• Fj - fixed cost of creating facility j.
• xij - variable for flow from facility j to
  customer i.
• Yj - binary variable that equals 1 if we create
  facility j, 0 otherwise
• sj - capacity of facility j.

45
Uncapacitated Facility Location Model Formulation
46
Capacitated Facility Location Model Formulation
47
Supply Chain Design Model

• The objective of this model is to determine the
  warehouse and plant configuration that minimizes
  total costs for production and distribution of
  multiple products.
• Based on Geoffrion and Graves, 1974,
  Multicommodity distribution system design by
  Benders decomposition, Management Science, v20,
  n5. (see Tech. Suppl., Ch. 13)
• Notation
• i - index for commodities
• j - index for plants
• k - index for warehouses
• l - index for customer zones

48
Supply Chain Design Model

• Notation (continued)
• Sij - production capacity for commodity i at
  plant j.
• Dil - demand for commodity i in customer zone l.
• - min and max total throughput for
  warehouse k.
• fk - fixed part of annual costs for owning and
  operating warehouse k.
• vk - variable unit cost of throughput for
  warehouse k.
• Cijkl - average unit cost of producing, handling,
  and shipping commodity i from plant j through
  warehouse k to customer zone l.
• Xijkl - amount of commodity i flowing from plant
  j through warehouse k to customer zone l.
• ykl - binary variable 1 if warehouse k serves
  customer zone l, 0 otherwise
• zk - binary variable 1 if warehouse k is open,
  0 otherwise.

49
Supply Chain Design Model Formulation
50
Network Planning

• Network planning refers to assessing or
  reassessing the configuration of facilities,
  commodities, and flows currently used to satisfy
  demand
• Network planning data checklist
• List of all products
• Customer, stocking point, and source point
  locations
• Demand by customer location
• Transportation rates
• Transit times, order transmittal times, and order
  fill rates
• Warehouse rates and costs
• Purchasing/production costs
• Shipment sizes by product
• Inventory levels by location, by product, control
  methods
• Order patterns by frequency, size, season,
  content
• Order processing costs and where they are incurred

51
Network Planning

• Data Checklist (continued)
• Capital cost
• Customer service goals
• Available equipment and facilities and their
  capacities
• Current distribution patterns (flows)
• Note that many of these are decision variables
• Accumulating these data usually results in
  improvements by uncovering anomalies
• We must decide our network design strategy
• Specify minimum service levels
• Specify shortage costs and minimize cost
• Levels of acceptable aggregation of demand
• Optimization vs. heuristic methods
• Which areas require the most accuracy and
  attention?

52
The Role of Information Systems in Network
Planning

• The past five years have seen an explosion in
  customized installations and implementations of
  supply chain and total resource planning software
  packages
• These packages, such as SAP and PeopleSoft,
  integrate all of the elements on the previous
  network planning data checklist
• These packages integrate human resources,
  accounting, finance, production (push and pull
  capable), marketing, and distribution (DRP)
  systems
• The common name for such systems is ERP
  (Enterprise Resource Planning), although several
  companies have systems dedicated to supply chain
  (production and distribution) planning and
  scheduling
• The goal of ERP systems is to have one integrated
  place for all vital corporate data.
• Many of the jobs in IE consulting these days
  focus on implementation and customization of
  these systems for manufacturing and service firms.