A SYSTEMS DEVELOPMENT LIFE CYCLE METHODOLOGY FOR AGENT BASED MODEL DEVELOPMENT

1

NKS 2006 WOLFRAMSCIENCE CONFERENCE

• A SYSTEMS DEVELOPMENT LIFE CYCLE METHODOLOGY FOR
  AGENT BASED MODEL DEVELOPMENT
• ROD SIPE NEW SCIENCE PARTNERS
• June 16, 2006

2
Introduction

• Rod Sipe
• Career Corporate IT
• 20 Years At Ernst Young
• Liaison to BiosGroup
• New Science Partners
• Dedicated to the Application of CAS to Business
• Application Concepts
• Project Management
• Methodologies

3
Applications In The Energy Industry
Planning Marketing Production Distribution
Predictive Modeling Gas/Elec. Market Convergence Commodity Trading Work Over Economics Alternative Path Models
Schedule Optimization Refinery Location Commodity Transport Cost Well Production And Maintenance Trucking And Pipeline Operations
Decision Support Drilling Portfolio Hedging, Financial Instruments Spark Spread Interrupt Recovery
4
Applications In The Energy Industry

• Examples
• Pipeline Schedule Optimization
• Well Work Over Support Equipment Forecasting
• Drilling Portfolio Optimization
• Oil Product Delivery Truck Routing
• Economic Optimization of Sales/Transportation
  Options

5
(No Transcript)
6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
Model Types, Description, Value and Users
Model Type Description Value User
Proof of Concept Proves Feasibility Funding Core Development Team
Baseline Model Establishes the Development Environment Closure on Basic Model Performance CDT, Subject Matter Experts
Refined Model Complete Enough to Begin Getting Business Answers Business Insight Through Iteration and Observation CDT, SME
Expert Interpretation Operated by Experts Who Interpret the Results Answer Specific Situational Questions SME
Decision Support Answers Pre-Specified, Recurring Questions Operational Decision Support Super Users
Integrated Production Model Plays a Role in Regularly Scheduled Business Activities Optimization of Production Process Line Personnel
16
Concepts of CAS In Business

• High Probability Scenarios
• Multiple Factors
• Frequent Iteration
• Random Disturbance
• Short Cycle Times
• Many Possible Solutions
• Practical Expectations
• Know Your Potential Business Value
• Optimization of What?
• Actionable Results
• What Do You Know?
• Business Process Models
• Entities and Attributes
• What Are The Key Factors
• How Do They Behave

17
Concepts of CAS In Business

• What Is Knowable
• Butterfly Effect
• Answers Vs. Tendencies
• Data Access Architecture
• Global View
• Business Processes
• Technical Fragmentation
• Research And Development, Not Systems Development
• A New Understanding
• Iteration
• The Right To Not Reach The Objective
• The Breakthrough Result
• Counter-Intuitive Solutions

18
OBSERVATIONS

• State of the Science
• -Stus Cambrian Explosion Model
• -Limited Modeling Vocabulary
• Attitudes
• -Assumptions
• -Expectations
• -Precedent
• -Status Quo

• Business Value
• -Counter Intuitive
• -Value Proposition
• -Lack of Self Knowledge
• -Lack of Data Access
• Methodology
• -RD vs. Systems Development
• -Systems Development Life Cycle Methodology
  Meets The Scientific Method

19
FOR MORE INFORMATION

• Rod Sipe
• rsipe_at_houston.rr.com
• 713 504 6331
• www.newsciencepartners.com