1 Chapter 12: Decision-Support Systems for Supply Chain Management Prepared by Zouxin.

1

Chapter 12: Decision-Support Systems for Supply Chain Management

Prepared by Zouxin

2

Contents

Case study1. Introduction2. The Challenges of Modeling3. Structure of Decision-Support Systems4. Supply Chain Decision-Support System5. Selecting a Supply Chain DSS6. Summary

3

Case study-Supply chain management smooths production flow Aerostructures Corp.’s

A manufacturer of wings and wing components. At present: Rhythm – A Supply chain management system from i2

Technologies, Inc.Benefit: level out work flow Saves $500,000 of inventory costs.

In the past: MRP-II system Shortage: Couldn't schedule any smaller jobs. Couldn't afford to let unfinished products sit around for too long bec

ause of 220 operations Difficult to order materials

By this chapter Goal of software What types of decision support tools should be chosen?

4

1. Introduction Problems in Supply chain management are not so rigid and w

ell defined that they can delegated entirely to computers.

DSSs are used from strategic problems (logistic network) to tactical problems (assignment of products to warehouse / factory)

DSS uses mathematical tools (Operations Research, Artificial Intelligence)

DSS uses statistical tools (Data mining) and data warehouses.

5

Framework for SCMS based on planning horizon.

1. Strategic network design

2. Supply chain master planning

3. Operational planning (demand planning, inventory management, production scheduling, transportation planning systems )

4. Operational execution (enterprise resource planning, customer relationship management, supplier relationship management, supply chain management and transportation systems)

6

2. The Challenges of ModelingModel is the heart of any DSSsMajor questions when modeling supply chains

What part of reality should be modeled? On the one hand, model should include enough detail to

represent reality. On the other hand, model should be simply enough to understand, manipulate, and solve.

“Model simple, think complicated”

What is the process of modeling?“Start with a simplified model and add complexity later ”

What level of data and detail is required?“Modeling needs drive data collection, not the other way

around”

7

3. Structure of Decision-Support Systems Three major components:

3.1 Input database and parameters Input database contains the basic information needed for decisi

on making. parameters and rules also included, such as desired service le

vel, restrictions, various constraints 3.2 Analytical tools

The data analysis usually Involves embedded knowledge of the problem, while also allowing the user to fine-tune certain parameters.

Analytical tools include operations research, artificial intelligence, cost calculators, simulation, flow analysis, etc.

3.3 Presentation tools Display the results of DSS analysis. Ex) GIS, Gantt charts

8

3.1 Input Data Input data is critical to the quality of the analysis.

Depending on the type of analysis, a DSS may require collecting information from various parts of a company.

Model and data validation is essential to ensure that the model and data are accurate enough.

The decision planning horizon affects the detail of the data required.

examples [E.12-1] Input data required for logistics network design [E.12-2] Input data required for supply chain master planning

9

3.2 Analytical Tools Common DSS analysis tools and techniques :

Queries simply by asking specific questions about the data.

Statistical analysis To determine trends and pattern in the data.

Data mining to look for “hidden” patterns, trends, and relationship in the data.

On-Line analytical process (OLAP) tools Provided an intuitive way to view corporate data OLAP tools aggregate data along common business dimensions

and let users navigate through the hierarchies and dimensions by drilling down, up, or across levels.

Calculators to facilitate specialized calculations such as accounting costs.

10

3.2 Analytical Tools Simulation

to help decision making in random or stochastic elements of a problem.

Artificial Intelligence Employed in the analysis of DSS input data. Expert system captures an expert’s knowledge in a database and

use it to solve problems. Mathematical Models and Algorithms

Exact algorithms find mathematically “the best possible solution” of a particular problem.

Heuristics algorithms provide good, but not optimal solution to the problems.

It is often useful if in addition to the solution, the heuristic provides an estimate of how far the heuristic solution is from the optimal solution.

11

The analytical tools used in practice are typically a hybrid of many tools.

Applications and analytical toolsproblems Tools used

marketing Query, statistics, data mining

routing Heuristics, exact algorithms

Production scheduling Simulation, heuristics, dispatch rules

Logistics network configuration Simulation, heuristics, exact algorithms

Mode selection Heuristics, exact algorithms

The table shows a number of problems and analytical tools that are appropriate for them

12

3.3 Presentation Tools

Geographic Information Systems GIS is an integrated computer mapping and spatial database manage

ment system that can provide geographically referenced data. GIS can be used in many areas, GIS can be applied in supply

chain management, such as Network analysis—transportation, telecommunications Site selection Routing Supply Chain Management

Geographic Information Systems

Presentation ToolsIntegrating Algorithm and GIS

13

Integrating Algorithm and GIS

Geographic data Attribute data

GIS engine/map

Network Solution strategy

Algorithms

A general framework for integrating algorithms and GIS

14

4. Supply Chain Decision-Support System

Logistics network design Supply chain master planning Operational planning systems Demand planning Inventory management Transportation planning Production scheduling Material requirements planning (MRP) Operational executing systems

15

4. Supply Chain Decision-Support System

Logistics network design Involves the determination of warehouse and factory locations

and the assignment of retailers to warehouses. Heuristic or exact algorithms are used to suggest

network designs. Supply chain master planning

Process of coordinating production, distribution strategies, and storage requirements to efficiently allocate supply chain resources.

It is very difficult to do a supply chain master planning manually and an optimization-based decision-support system is needed.

16

Production Planning

Supply chain Tactical model and planning

Demand planning/Order

fulfillment

Detailed

production

schedule

Supply chain

master plan

Profit by market

and product

Demand forest

demand shaping

Feasibility Cost/profit Service level

The extended supply chain: from manufacturing to order fulfillment

17

4. Supply Chain Decision-Support System Operational planning systems

Includes different types of systems, ranging from demand planning tools to tools that assist with the details of production and sourcing strategies.

Demand planning Demand planning tools allow supply chain executives to

apply two different processes Demand forecast: long-term estimates of expected demand. Demand Shaping: A process in which the firm determines the impa

ct of various marketing plans such as promotion, pricing discounts, rebates, new product introduction, and product withdrawal on demand forecasts.

Inventory management To determine the levels of inventory, safety stock levels, to keep in ea

ch location in each period. In almost all cases, DSS apply a heuristic algorithm to generate sug

gested policies.

18

4. Supply Chain Decision-Support System Transportation planning

Involves the dispatching of a company's own fleet and decisions regarding selection of commercial carrier on certain routes.

Static and dynamic system (for example: telephone repair crews)

Production scheduling Production scheduling DSSs purpose manufacturing sequences and

schedule, given a series of products to make, information about their production processes, and due dates for the product .

Usually use artificial intelligence and mathematical and simulation techniques to develop schedules.

Material requirements planning (MRP) Use a product’s bill of materials and component lead times to plan whe

n manufacturing of a particular product should begin.

19

Operational executing systems These are real-time systems that allow executives to

run their business efficiently. DSSs can provide three levels of sophistication

Available to promise (ATP): firm can consider finished goods inventory as well as work in process to make a decision.

Capable to promise (CTP); firm can check components and materials availability to make a decision.

Profitable to promise (PTP): firm considers capability and profitability of completing an order

20

5. Selecting a Supply Chain DSS Issues considered in evaluating a particular DSS:

The scope of the problem, including the planning horizon The data required by DSS Analysis requirements – optimization, heuristics, simulation,

and computational speed needed. The system’s ability to generate a variety of solutions The presentation requirements Compatibility and integration with existing systems Hardware and software system requirements. The overall price Complementary systems

21

6. SummaryThe major trends in supply chain DSSs

1. Integration with and between ERP systems. DSSs will be easier to integrate with ERP systems through

standard interfaces.

2. Improved optimization Many DSSs lack a true optimization capability. Most existing supply

chain master planning and MRP systems do not optimize at all and in many cases do not take capacities into account.

3. Impact of standards. Many DSSs are not compatible and difficult to integrate. Strategic partnering forces the various partners to define standards.

4. Improved collaboration. Collaboration can enhance production planning, inventory management, an

d other supply chain process.