Title: Partner Selection in End to End Resource Planning: An SOA and Data Mining Based Approach
1Partner Selection in End to End Resource
Planning An SOA and Data Mining Based Approach
www.oasis-open.org
MIN NI, PHD Post-doctor research
scientist Digital China Post-doctor station
2www.oasis-open.org
Outline
- EERP
- Background
- Real-time Value Chain Optimization
- Design Implementation strategy
- Supplier selection
- Challenge of the problem
- Selection strategy
- QFD Data Mining
- Supplier selection implementation on EERP
- Conclusion
3www.oasis-open.org
Outline
- EERP
- Background
- Real-time Value Chain Optimization
- Design Implementation strategy
- Supplier selection
- Challenge of the problem
- Selection strategy
- QFD Data Mining
- Supplier selection implementation on EERP
- Conclusion
4Changing of business environment
- Business environment is tougher
- Rapid changing of market landscape
- Globalization
- Companies want to maximize their business
results - Revenues up
- profits up
- costs down
- Sustaining profit growth becomes critical
5State of the enterprise
- Crisis of Complexity
- No green-field development
- New applications must integrate with existing
heterogeneous legacy systems - IT Must Deliver Business Agility
- Support new business opportunities
- Leverage new technologies
6Changes in IT Architecture
- Business management pushes for better IT
utilization and greater ROI - Re-Use of IT assets, integration of B2X and ERP.
- Merger and acquisition have become a fact of life
- Integration of historically separate systems
across enterprise borders - New systems must be developed with (current or
future) heterogeneity in mind - Results driven.
- Internet everywhere has created a set of new
business models - Integration anywhere and anytime is possible.
7Interoperability via SOA
- Data, logic and infrastructure accessed by
routing XML-based messages (Web Service)
throughout Internet - Location is meaningless
- Multiples Services suppliers
- Ubiquitous computing
- Preserves and connects existing systems
- Using the Right Software Architecture brings
exponential productivity gains
8SOA in Action (example)
9www.oasis-open.org
Outline
- EERP
- Background
- Real-time Value Chain Optimization
- Design Implementation strategy
- Supplier selection
- Challenge of the problem
- Selection strategy
- QFD Data Mining
- Supplier selection implementation on EERP
- Conclusion
10SOA has proven
- Data level
- Lower cost, and more efficient way to integrate
application across the enterprises. - Application level
- Development with Lower entry barrier, easier way
to collaborate, and faster development cycle.
11Next step of SOA
- Business level
- How to coordinate different services in business
context and achieve the best business model? - How to generate the most efficient business
process automatically and intelligently? - The answer is EERP
- End to End Resource Planning
12EERP concept resource planning
- With SOA and open standard, we can now help
enterprises to improve their profits and sustain
the profits, we introduce EERP - End to end resource planning, optimize the
resources along an specified value chain.
13What does EERP do?
- What is it?
- A new way to look at enterprise IT application
- Integration inside and outside of a enterprise
border. - A new software architecture and platform
- a collection of Web Services that perform various
enterprise functions, data exchange, process
integration, collaboration and other relevant
work. - A self-healing, intelligent, and reliable service
choreograph platform (SoftGrid)
14Real-time Value Chain Optimization
15www.oasis-open.org
Outline
- EERP
- Background
- Real-time Value Chain Optimization
- Design Implementation strategy
- Supplier selection
- Challenge of the problem
- Selection strategy
- QFD Data Mining
- Supplier selection implementation on EERP
- Conclusion
16EERP top-down service design and deployment
strategy
Step 1
Step 3
Step 5
Step 7
Business ontology definition
Service oriented analysis
Service implementation
deploy service to EERP platform
Step 2
Step 6
Step 4
Service testing
Business process discovery definition
Service oriented design
17EERP top-down business process orchestration
strategy
Step 1
Step 3
Step 5
Step 7
Business ontology definition
Service discovery
Service orchestration
execute QoS auditing
Step 6
Step 8
Step 2
Step 4
Service (business process) execution
service QoS recording assessment
Business goal Investigation/ definition
QoS deployment calculation
18www.oasis-open.org
Outline
- EERP
- Background
- Real-time Value Chain Optimization
- Design Implementation strategy
- Supplier selection
- Challenge of the problem
- Selection strategy
- QFD Data Mining
- Supplier selection implementation on EERP
- Conclusion
19Value Chain Integration
20Integrating the Supply Chain Value Chain
- A Supply Chain transforms into an integrated
Value Chain when it.. - Extends the chain all the way from all levels of
suppliers to customers. - Integrates the back-office operations with those
of the front office. - Becomes highly customer-centric, focusing on
demand generation and customer service. - Is proactively designed by chain members to
compete as an extended enterprise. - Seeks to optimize the value added by information
and utility-enhancing services.
21Challenge of the problemto build value chain
agilely
- How to identify voice of customer', and satisfy
them - The complexity brought by the enormous amount of
suppliers (over 1000 suppliers for an automobile
manufacturer) - The efficiency of decision-making
- The challenge of integrating enterprise and its
suppliers
22www.oasis-open.org
Outline
- EERP
- Background
- Real-time Value Chain Optimization
- Design Implementation strategy
- Supplier selection
- Challenge of the problem
- Selection strategy
- QFD Data Mining
- Supplier selection implementation on EERP
- Conclusion
23Selection strategy
- Key Success Factors (QoS factors)
- -Quality
- -Cost
- -Delivery time
- -Innovation
- Continuous Improvement
- Total Value-Chain Analysis
- Dual External / Internal Optimization Focus
Customer Satisfaction
24www.oasis-open.org
Outline
- EERP
- Background
- Real-time Value Chain Optimization
- Design Implementation strategy
- Supplier selection
- Challenge of the problem
- Selection strategy
- QFD Data Mining
- Supplier selection implementation on EERP
- Conclusion
25(QFD) Quality Function Deployment Process
- How to capture customers requirements
- QFD (aka House of Quality) was first
implemented by Yoji Akao integrate the voice of
the customer into the design and delivery of
products.
26The House of Quality
5. Technical Correlations
3. Technical Characteristics
4. Customer Requirements -Technical
Characteristics correlation matrix
2. Product planning matrix
1. Customer Requirements
6. Goal constraints
27QFD Suppliers Selection
- Draft a list of supplier requirements that
contribute to the fulfillment of customers needs. - Formulate evaluation criteria for suppliers.
- Determine the importance (weight) of the
criteria. - Rank your suppliers, and make appropriate
selection that will accommodate customer needs.
28Extended House of Quality reflecting candidate
suppliers
29Data Mining in supplier selection
- By data mining, frequent fault patterns under
different environments and different use
patterns, critical requirements can be found - As mining result, frequent fault patterns can be
used to build a components correlation matrix - Manufacturers can perform competitive product
analysis based on information from customer
profiles and the market in order to determine
competitive products that greatly affect the
concerned product.
30www.oasis-open.org
Outline
- EERP
- Background
- Real-time Value Chain Optimization
- Design Implementation strategy
- Supplier selection
- Challenge of the problem
- Selection strategy
- QFD Data Mining
- Supplier selection implementation on EERP
- Conclusion
31An SOA-based System for supplier selection
Why use SOA? Data/Services are distributed/heterog
eneous Why choose EERP? To build real-time
value chain and perform optimization
32An SOA-based System for supplier selection based
on extended QFD and DM
Services implemented Data object
services Mining services Decision-making
services Optimizing services
33System implementation
- System requirements
- Operation System Windows NT/2000/XP/2003, LINUX,
UNIX - DBMS IBM DB2, Microsoft SQL Server, Sybase,
Oracle, Informix and DBMS that support JDBC/ODBC - Application server WebLogic, websphere, etc
34System interface
35www.oasis-open.org
Outline
- EERP
- Background
- Real-time Value Chain Optimization
- Design Implementation strategy
- Supplier selection
- Challenge of the problem
- Selection strategy
- QFD Data Mining
- Supplier selection implementation on EERP
- Conclusion
36Conclusion
- SOA provides the possibility and flexibility to
integrate heterogeneous systems - EERP use information gathered from the products
full life-cycle to perform real-time value chain
optimization, which is needed in real-world
business scenarios - EERP will lead the trend of enterprise computing
- Data mining can discover useful and interesting
knowledge in business context.
37Thanks!