The Role of Statistical Engineering in Creating Solutions

  • Slides: 28
Download presentation
The Role of Statistical Engineering in Creating Solutions for Complex Opportunities GEOFF VINING 1

The Role of Statistical Engineering in Creating Solutions for Complex Opportunities GEOFF VINING 1 M AY 2020 DATAWORKS 2020

Introduction to The Session My Job: ◦ Provide Big Picture ◦ Context for Statistical

Introduction to The Session My Job: ◦ Provide Big Picture ◦ Context for Statistical Engineering (SE) Adam: NIST ◦ Wonderful Example ◦ Great Illustration of the Phases Jim: Do. D Applications ◦ Outstanding Job Showing Applicability to Do. D Opportunities ◦ Wonderful Job Integrating Tools within SE Framework

Outline Statistical Engineering – Big Picture ◦ Scientific Method ◦ Systems – Statistical Thinking

Outline Statistical Engineering – Big Picture ◦ Scientific Method ◦ Systems – Statistical Thinking Phases to a Statistical Engineering Project Core Processes and Over-Arching Competencies Experiences Teaching SE This Semester Statistical Engineering’s Enduring Value

Big Picture Major Opportunities ◦ Unstructured – Mess! ◦ Usually Large ◦ Require Data

Big Picture Major Opportunities ◦ Unstructured – Mess! ◦ Usually Large ◦ Require Data Secrets to Success: ◦ Subject Matter Expertise for the Specific Problem! ◦ Tool Sets ◦ Strategical and Tactical Deployment

Basic Foundations Scientific Method Systems Thinking Structured, but Not Rigid, Approach “Adapt and Overcome”

Basic Foundations Scientific Method Systems Thinking Structured, but Not Rigid, Approach “Adapt and Overcome” Learn from the Journey

Scientific Method Sequential Learning Strategy! ◦ ◦ ◦ Understand the Real Problem at Hand

Scientific Method Sequential Learning Strategy! ◦ ◦ ◦ Understand the Real Problem at Hand Define the Problem Discover Solutions Develop a theory Test theory using data Modify theory as necessary Strong Need for Interdisciplinary Collaboration

Scientific Method Data Are the Keys to Successful Application ◦ Data collection ◦ Data

Scientific Method Data Are the Keys to Successful Application ◦ Data collection ◦ Data analysis ◦ Data Interpretation Quality Engineering/Industrial Statistics Are “Handmaiden” ◦ Subject Matter Insights Must Guide Everything! ◦ Data Analysis Must Follow ◦ All Solutions Require Subject Matter Validation/Verification Essential for Solving Complex Problems.

Systems Approach Strategy for Success: ◦ ◦ Right Tool Right Job Right Time Correctly

Systems Approach Strategy for Success: ◦ ◦ Right Tool Right Job Right Time Correctly Applied Tools Are Not the Focus of Statistical Engineering! Sustainable Solutions Are! Tools Are Not Solutions!

Schematic Diagram of Statistical Engineering People True SMEs from Crit. Disc. Team Building Collaboration

Schematic Diagram of Statistical Engineering People True SMEs from Crit. Disc. Team Building Collaboration Statistical Engineering: Strategy and Tactics Stat, DS, Bus Anal ISE, OR, M&S Tool Sets Org & Behav. Psych Better, Faster Solutions to Complex Problems

Phases of Statistical Engineering (1) Identify Problem (2) Provide Structure (3) Understand Context ◦

Phases of Statistical Engineering (1) Identify Problem (2) Provide Structure (3) Understand Context ◦ ◦ Science Business Personnel Politics Iterate (1)-(3) as Necessary

Phases of Statistical Engineering (4) Develop the Solution Strategy (5) Develop and Execute Tactics

Phases of Statistical Engineering (4) Develop the Solution Strategy (5) Develop and Execute Tactics (6) Deploy Final Solution Iterate (4) and (5) as Necessary ◦ ◦ Failures Are Important “Make the Call” Earlier Rather than Later Investigate Thoroughly Learn and Document

After the Project Document the Process as a Case Study Summarize ◦ Worked Well

After the Project Document the Process as a Case Study Summarize ◦ Worked Well ◦ Did Not Work ◦ Recommendations for Next Time “Publish” the Final Document

Core Processes – Critical Bodies of Tools Data Acquisition Data Exploration Modeling ◦ Traditional

Core Processes – Critical Bodies of Tools Data Acquisition Data Exploration Modeling ◦ Traditional Statistical Methodologies ◦ Modern Analytics (Big Data) ◦ Modeling & Simulation

Core Processes – Critical Bodies of Tools Inference to the Process/Problem Evaluation and Trial

Core Processes – Critical Bodies of Tools Inference to the Process/Problem Evaluation and Trial Deployment of Tentative Solutions Deployment of Final, Sustainable Solution

Overarching Competencies Data Visualization Optimization Project Management Teamwork Organizational Culture

Overarching Competencies Data Visualization Optimization Project Management Teamwork Organizational Culture

Experiences Teaching SE This Semester First Time Taught Anywhere Eight Students ◦ Undergraduate: 1

Experiences Teaching SE This Semester First Time Taught Anywhere Eight Students ◦ Undergraduate: 1 – STAT and 3 – ISE (2 from Argentina) ◦ Graduate: 1 – STAT and 3 – ISE Only 1 Born in the US Only One with Some Experience in Working in Teams (UG)

Experiences Teaching SE This Semester Syllabus ◦ ◦ ◦ Big Picture of Statistical Engineering

Experiences Teaching SE This Semester Syllabus ◦ ◦ ◦ Big Picture of Statistical Engineering Intro to Team Building Overview and Details: Phases Overview and Details: Core Processes Overview and Details: Overarching Competencies Student Proposed Projects to Illustrate

First Major Project Define Parameters for Major Project Instructor Constraints: ◦ Sufficiently Complex; Meaningful

First Major Project Define Parameters for Major Project Instructor Constraints: ◦ Sufficiently Complex; Meaningful Progress by End of Term Basic Goals: ◦ Introduction: Collaboration, Bringing Structure to an Opportunity to Introduce Soft Tools ◦ NGT, Affinity Diagrams, Multi-Voting, Fishbone Diagrams

Second Major Project Work Together to Identify Appropriate Project Opportunity to Apply Soft Tools

Second Major Project Work Together to Identify Appropriate Project Opportunity to Apply Soft Tools Introduced ◦ Environmental Scans ◦ Gap Analysis ◦ Project Charters

Second Major Project Learned the Art of Compromise Identified Project: Try to Work with

Second Major Project Learned the Art of Compromise Identified Project: Try to Work with Blacksburg Transit Timing: Just Before Spring Break!

Basic Foundations of SE Strategy Scientific Method Systems Thinking Structured, but Not Rigid, Approach

Basic Foundations of SE Strategy Scientific Method Systems Thinking Structured, but Not Rigid, Approach “Adapt and Overcome” Learn from the Journey

Third Major Project: Socially Determined Non-Profit Organization Uses Analytics to Support Policy Decisions Specific

Third Major Project: Socially Determined Non-Profit Organization Uses Analytics to Support Policy Decisions Specific Project: ◦ Support Process to Advise Governor of MD ◦ Allocate Resources (Federal Funding) ◦ “Optimally” Mitigate Effects COVi. D-Impact ◦ Must Consider Full Effects of Virus on MD

Third Major Project: Socially Determined Weekly Zoom Interactions with Client Goals: ◦ ◦ ◦

Third Major Project: Socially Determined Weekly Zoom Interactions with Client Goals: ◦ ◦ ◦ Provide New Ideas on the Structure of Problem Find New Data Sources Suggest New Indices of Virus Impact Develop Interesting Database as Proof of Concept Time Permitting, Initial Analysis of Database

Reflection Definitely Not Original Plan Critical Factors to Success: ◦ Students Learned Early How

Reflection Definitely Not Original Plan Critical Factors to Success: ◦ Students Learned Early How to Collaborate as a Team! ◦ Adapted Well to Remote Meetings (Two Members in Argentina) ◦ Enthusiastic Participation of the Client ◦ Fred Faltin! “Adapt and Overcome”: Far Better Student Experience

The Journey There Is a Well-Defined Road Map However, Each Journey Is Unique Success

The Journey There Is a Well-Defined Road Map However, Each Journey Is Unique Success Requires: ◦ ◦ ◦ Teamwork/Collaboration Proper Blend of Skills – Especially Subject Matter Experts Proper Understanding of the Tools and Their Roles Proper Understanding How to Deploy the Tools Strategically Strong Leadership!

Enduring Value of Statistical Engineering Success Requires Providing Value to Organizations Addressing Opportunities Provides

Enduring Value of Statistical Engineering Success Requires Providing Value to Organizations Addressing Opportunities Provides Great Value Current State: Focus on Limited Number of Tools Future State: How Do We Use the Tools Most Effectively ◦ ◦ Most Understand the Full Range of Tools Required Most Understand Best Practices for Using the Tools Must Understand How to Deploy the Tools Strategically Must Understand Leadership in the Broad Sense The Future Is a Journey!

The Journey: ISEA New Professional Society Focus: The Emerging Discipline of Statistical Engineering Business

The Journey: ISEA New Professional Society Focus: The Emerging Discipline of Statistical Engineering Business Model Based on ENBIS ◦ Web Based ◦ Free Individual Memberships ◦ Paid Organizational Memberships Website: isea-change. org

Many thanks!

Many thanks!