Analytical Modeling of Logistics for

Reconfigurable Supply Chain in Mass

Customization of Vehicles

AUTHOR: Charu Chandra

Department of Industrial and Manufacturing Systems
Engineering

PARTNER: Mathew Zaluzec

Matthias Baumann
Ford Motor Company

Robert Hildebrand
Institute of Advanced Vehicle Systems

ASSISTANTS: Ruzanna Marukyan

Armen Tumanyan
Department of Industrial and Manufacturing Systems
Engineering

1
Background

 Efficiency of logistics operations has

major impact on mass customization
 Knowledge assumes a critical role in
company’s success, and, therefore
must be captured, organized and
utilized effectively

2
Objectives
• The main objective of the project is to analyze reconfiguration of supply
chain dealing with issues related to overseas production facilities:
– Local vs. distributed system
– Local suppliers’ selection
– Dynamic demand scenarios
– Robustness of the configuration to disruptive events
• Objective of the current phase of the project is to represent an
information support system (ISS) for a supply chain (SC) comprising:
 SC system taxonomic representation of SC structure, operation and
control
 Process models for explicit and exact representation of SC
 SC knowledge base organization
 Web application development for SC management (SCM) components

3
Approach
Technical Approach
 Development of an integrated framework to design,
model, and evaluate strategic decisions through
internet-based technologies.
 A five-step design and implementation approach is
proposed:
 SC Systems taxonomy development
 Process Model construction
 Database Design
 Ontology construction
 Development of integrated web application

4
Approach
Information system components I
Kernel SC Description

SC System
Taxonomy

SC Problem Ontology
Taxonomy

Process Model Repository

Construction

ERP Object-
Relational
Applications
Database Design
5
Approach
Information system components II
• Kernel describes SC characteristics based on a
Product-Process-Resource (PPR) model
• Taxonomy provides a unified theoretical foundation of
SC system and its domains
• Problem taxonomy formalizes SC problems
• Process model represents configurations
• Ontology bridges domain analysis (taxonomy) and
application system construction

6
 Approach
Taxonomy development stages

Systematics
Manufacturing differences based on System theory

Taxonomic presentation
Identify Manufacturing system boundaries,
Identify attributes for manufacturing system

Classification
Develop system or schema based on taxonomic proposals,
Collect data and develop groups of SC systems

Taxa
Groups of manufacturing systems

7
 Approach
Supply chain process model
configurations
P1 P3

PU1 P1, P2
• Standard relationships
P2 R1 PU3 between units are
PU2 identified

P1
• Process model
PU1 development rules are
P2
R1
P1, P2
PU4 elaborated
P3 PU2
P1 P4
P2
P3, P4
R2 PU5 PU1 R1 PU2
P4 P2
P3
PU3

8
Approach
Data Base model
 Specially designed data base model provides
flexible means for representing data relevant
to SC modeling
 Data base is used to store and to
disseminate:
 a standardized representation of the SC modeling
problem
 SC modeling results
 accumulated modeling knowledge
 Database reflects structure of PPR model.
9
Approach
Ontology modeling components
Concepts are notions describing domain entities, derived
Concepts
from system taxonomy.

Relations represent interactions between concepts of the

Relations domain. Relationships can of two types: Whole-Part and Slots.

Information model attributes are characteristics of the problem

Attributes solving environment.

Axioms are rules for the particular domain. They can include
Axioms equations or global constraints.

Meta-data can include any useful information that describes

Meta-Data the problem, or helps the user to understand the main
components: Concepts or Axioms.
10
 Application
Information support system
application
 A company locates manufacturing facilities in external market
 Production of customizable products
 Continuous improvement of SC configuration
 Selection of local suppliers
PS1
Plastic Interior
PS2
Design 1
Design 1
750,1000
MS1 Engine & other
overseas Assembly Final
Distribution
supplies assembly
MS2 Design 2 Design 2
Design 1
250,1250
MS3 Metal
Design 2
Exterior
MS4

MS5
Local suppliers

11
 Results
Software architecture

Information system provides

input data to the decision
modeling system

Legend:
– Information System Module
– Data Flow
– Cross-Application Interface

12
Results
Top level supply chain system
taxonomy

13
 Results
Supply Chain Problems Taxonomy: PPR
model taxonomy
Taxonomy of Product, Process, and Resources model

Ordering Policy
Product Production policy Resource
Delivery policy
Demand rule Process
- Cost Backlog penalty
- Time
Fixed resource costs
- Number of Product - Transformation Variable resource costs
- Selling Price - Customer Relationship Management Process
- Processing batch size - Customer Service Management
Demand - Customer Order Fulfillment
- Customer Demand - Demand Management Process
- Demand constraint - Product Development and Commercialization
- Procurement Process
- Manufacturing Flow Management Process
Cost Time
- Product cost Flows Human Marketing Organizational
Product development lead
- Holding cost
time Production Goodwill Training system
- Setup cost
- Marketing time Managers Brand Corporate culture
- Direct manufacturing cost
- Sales time Material flow Scientists Market share Supplier
- Indirect manufacturing cost
- Order processing time Information flow Machine operators relationships
- Inventory holding cost
- Product design time Product flow
- Outbound transportation cost
- Inventory control time Financial flow
- Inbound transportation cost
- Production planning time Demand flow
- Row material cost
- Material ordering time Order fulfillment Physical Legal
- Unit holding cost at particular - Material receiving time flow
location per time unit Copyright
Manufacturing Lead Time Plants
- Fixed ordering cost Contracts
- Storage time - Interest rate constraint Inventories
Patents
- Unit shipment cost per time unit - Average time - Changes in working capital
- Unit inbounding cost per time unit - Move time - Retained earnings balance
- Queue time Supply revenue
- Unit outbounding cost per time unit - Income equations
- Stock out time Minimum inventory level Supply availability
- Unit processing cost per time unit - Minimum earning constraint Opening inventory level Sourcing constraint
- Batch shipment cost - Setup time - Minimum dividend requirement
- Run time Inventory capacity
- Batch inbounding cost - Tax credit equations Distribution center
- Down time - Inventory balance equations
- Batch processing cost - Rework time constraints
- Batch outbounding cost - Minimum taxes paid Production capacity

14
Results
Process data model

15
Results
Database prototype for Supply Chain
Structures

16
 Results
Configuration problem knowledge
module (ontology)
Supply Chain Configuration

Input Agent Functions Output

Materials
Customer Structure Products
-Supplier_ID:int
-Material_ID:int -Customer_ID:int -Supplier_ID:int
-PU_ID:int -Custmer_name:int -Product_Name:int
-Material_Name:int Production Unit -Product_ID:int
-PU_ID:int
-PU_ID:int Product Structure
-PU_Name:int
-Product_ID:int
Material Attributes -Product_Name:int

-Material_Shipment_Costs:int
-Materials_Capacity:int
-Quantity_of_Materials:int Product attributes
Production Unit Attributes
-Material_Costs:int Components
-Invemtory_of_Product:int
-PU_Open_Indicator:int -Quantity_of_Products_Shipped:int
-Capacity:int -Material_Name:int
-Material_ID:int -Quantity_of_Product_Produced:int
-DC_Open_Cost:int -Quantity_Shipped:int
Customer Attributes -PU_Open_Cost:int -Quantity_Product_Cost:int
-Inventory_Cost:int
-Demand:int -Capacity_Cost:int

17
 Configuration results
Structural data (process model) and logical data (taxonomy) are feed to
the decision modeling system to obtain quantitative configuration results
PS1, 1800,1800
1800 Interior,
2420 Optimal suppliers
PS2,
2420 2420,2420
Interior,
2420
2420,2420
Large suppliers
PS2, 620,620 2420,2420
620
2420,2420 2420,2420 1120,1120 760,1480
760,1480
1120,1120
Engine & other Engine & other Final
Final Assembly,
MS1, 1020,1020 overseas Assembly, overseas assembly,
1020 1120,1300
assembly, 1120,1300
supplies 1120,1300 supplies 1120,1300
1300,1300 1120,1120 1300,1300 360,2220
MS4, 1120,1120 360,2220
1200 I(Design 1)=360, I(Design 1)=360,
1200,1200
I(Design 2)=920 3720,3720 1300.1300 I(Design 2)=920
Exterior, MS2, Exterior,
1300.1300 3720 1120,1300
MS5, 1500,1500
1120,1300
1500

Quantitative comparison of configurations

5 80
Optimal suppliers

Delivery Time
4 76
3 Large suppliers 72
Costs

2 68
1 64
0 60
Costs according to Costs according to Delivery times
strategic optimization operational planning according to simulation

18
 Impact
 Educational Impact: SC management graduate
course curriculum
 Models and methods for SC integration and synchronization
 Enterprise information system design, modeling, and
integration

 Industrial Impact:
 Accumulation of SC management knowledge
 Building dynamic decision making environment
 Application of generic approaches to other industrial
domains

19
 Future Work
Application Theory
 Problem Taxonomy object-
 Web-based implementation
of the SC information system oriented modeling environment
and knowledge library development
 Multi-level process model
 Integration with ERP
construction,
operational modules  Experimenting with key

elements of the taxonomy for

mass customization,
 Developing theoretical

constructs on the basis of

above experimentation
 Ontological Representation of a

SC
20