Data-Driven Optimization

Similar documents
ANALYTICS CENTER LEARNING PROGRAM

JOB DESCRIPTION DEPARTMENT: JOB TITLE: Senior Data Analyst JOB GRADE: JOB CODE : REPORTS TO: Senior Manager : Supply Chain Centre

CoolaData Predictive Analytics

MS, Tepper School of Business, Carnegie Mellon University, Major: Operations Research

PROGRAM DIRECTOR: Arthur O Connor Contact: URL : THE PROGRAM Careers in Data Analytics Admissions Criteria CURRICULUM Program Requirements

Graduate Program in Transportation Science

one Introduction chapter OVERVIEW CHAPTER

Industrial and Systems Engineering Master of Science Program Data Analytics and Optimization

Role of Social Networking in Marketing using Data Mining

Data-Driven Decisions: Role of Operations Research in Business Analytics

Part VIII: ecrm (Customer Relationship Management)

How To Get A Masters Degree In Logistics And Supply Chain Management

Information and Decision Sciences (IDS)

Attaining Supply Chain Analytical Literacy. Sr. Director of Analytics Wal Mart (Bentonville, AR)

Interpreting Web Analytics Data

Sanjeev Kumar. contribute

Business Analytics with. Management Science. Models and Methods. Arben Asllani Professor of Business Analytics, University of Tennessee at Chattanooga

AdTheorent s. The Intelligent Solution for Real-time Predictive Technology in Mobile Advertising. The Intelligent Impression TM

High-Performance Analytics

Study Plan for the Master Degree In Industrial Engineering / Management. (Thesis Track)

The Complete Guide to Unlocking (Not-Provided) Data in Google Analytics

Operations Research Department of Integrated Systems Engineering The Ohio State University

STOCHASTIC ANALYTICS: increasing confidence in business decisions

Data Isn't Everything

The Master s Degree with Thesis Course Descriptions in Industrial Engineering

APPLICATION OF INTELLIGENT METHODS IN COMMERCIAL WEBSITE MARKETING STRATEGIES DEVELOPMENT

Building a Database to Predict Customer Needs

Chapter 7: Data Mining

Doctor of Philosophy in Computer Science

SAS for the Department of Business Studies, ASB

Voice. listen, understand and respond. enherent. wish, choice, or opinion. openly or formally expressed. May Merriam Webster.

SEM for successful campaign management

Database Marketing simplified through Data Mining

TEXT ANALYTICS INTEGRATION

Optimal Scheduling for Dependent Details Processing Using MS Excel Solver

COPYRIGHTED MATERIAL. Contents. List of Figures. Acknowledgments

Principles of Data Mining by Hand&Mannila&Smyth

Applied Analytics in a World of Big Data. Business Intelligence and Analytics (BI&A) Course #: BIA 686. Catalog Description:

Data Technologies for Quantitative Finance

Beyond listening Driving better decisions with business intelligence from social sources

MSCA Introduction to Statistical Concepts

Optimization applications in finance, securities, banking and insurance

GLOBAL PREMIUM A2P (MT) & P2A (MO) SMS/MMS MESSAGING MARKET SIZE & FORECAST ( )

How Organisations Are Using Data Mining Techniques To Gain a Competitive Advantage John Spooner SAS UK

The Masters of Science in Information Systems & Technology

IBM Social Media Analytics

Master of Science in Computer Science

Executive Master's in Business Administration Program

Digging for Gold: Business Usage for Data Mining Kim Foster, CoreTech Consulting Group, Inc., King of Prussia, PA

Master of Mathematical Finance: Course Descriptions

KnowledgeSTUDIO HIGH-PERFORMANCE PREDICTIVE ANALYTICS USING ADVANCED MODELING TECHNIQUES

Management MBA (MGMTMBA)

Big Data Hope or Hype?

Flexible Core FC (Total 12 Credits) Analytical Skills People Skills PS. Credits

Operations Management OPM-301-TE

WordStream Helps New Agency Indulge in PPC Advertising

Technical Elective Tracks

Research Article Determination of Pavement Rehabilitation Activities through a Permutation Algorithm

Azure Machine Learning, SQL Data Mining and R

Howe School of Technology Management. Applied Analytics in a World of Big Data. Business Intelligence and Analytics (BI&A) Proposed Course #: BIA 686

Analytics as a Service Overview

Master of Science in Marketing Analytics (MSMA)

AACSB Annual Assessment Report For

Implementing Large-Scale Autonomic Server Monitoring Using Process Query Systems. Christopher Roblee Vincent Berk George Cybenko

Towards a Thriving Data Economy: Open Data, Big Data, and Data Ecosystems

Predictive Analytics

MSD Supply Chain Programme Strategy Workshop

Data Mining with SAS. Mathias Lanner Copyright 2010 SAS Institute Inc. All rights reserved.

An Overview of Knowledge Discovery Database and Data mining Techniques

Evolution of Business Intelligence in the Digital Age Dr. Gautam Shroff Vice President and Chief Scientist, TCS Research

Intrusion Detection: Game Theory, Stochastic Processes and Data Mining

MEETING THE CHALLENGES OF COMPLEXITY AND SCALE FOR MANUFACTURING WORKFLOWS

CHAPTER 2 PAVEMENT MANAGEMENT SYSTEM

Flexible Core FC (Total 12 credits) Analytical Skills SS. Credits

Big Data Optimization at SAS

GAMS Productivity - Performance - Reliability

Implementing Data Models and Reports with Microsoft SQL Server

Web analytics: Data Collected via the Internet

Declared as Deemed-to-be University u/s 3 of the UGC Act,1956. For enrolment, send in your request to connect@analyticsindia.org

TRANSFORMING LIFE SCIENCES THROUGH ENTERPRISE ANALYTICS

2015 Workshops for Professors

Business Intelligence Solutions for Gaming and Hospitality

The Definitive Guide to Google AdWords

Resource Allocation Modeling Techniques Applied to Air Force Crew Scheduling Problem

Business Analytics Syllabus

Alcatel-Lucent 8920 Service Quality Manager for IPTV Business Intelligence

Transcription:

Data-Driven Optimization John Turner Assistant Professor The Paul Merage School of Business University of California, Irvine October 24, 2014: UCI Data Science Initiative

Analytics Unlocking the Potential of Big Data è Descriptive What is happening? Querying, Reporting, Data Capturing, Filtering & Analysis è Predictive What will likely happen? Statistical Methods (Regression), Forecasting & Data Mining è Prescriptive What should we do? Optimization, Simulation, Quantitative Models John Turner - UCI Data Science Initiative 2

John Turner - UCI Data Science Initiative 3 Optimization min f(x) @ i=1..m@ i=1..p@ x X s.t. g i (x) 0, h i (x)=0,

Applica'ons for Op'miza'on chip design logistics supply chain management portfolio optimization scheduling network design air traffic routing timetabling marketing campaign design production planning

John Turner - UCI Data Science Initiative 5 Problems, Formula'ons, and Instances Define Problem Algebraic Manipulation Construct Mathematical Model (Formulation) Solver (CPLEX, Gurobi, KNITRO, etc.) Solve one or more Instances Communicate Results / Automate the Process

John Turner - UCI Data Science Initiative 6 Computational Challenges (x) @.m@.p@ x X Nonconvex objective s.t. g i (x) 0, h i (x)=0, Too many constraints Nonlinear constraints Too many variables Nonconvex domain (e.g., discrete X) à Large-Scale Instances Require Specialized Decomposition or Approximation Schemes

Two Examples 1. Planning Online Advertising Turner, J. 2012. The Planning of Guaranteed Targeted Display Advertising. Operations Research, 60(1) 18--33. 2. Locating Trauma Centers at Nation-Scale Cho, S-H., H. Jang, T. Lee, J. Turner. 2014. Simultaneous Location of Trauma Centers and Helicopters for Emergency Medical Service Planning. Operations Research, 62(4) 751--771. John Turner - UCI Data Science Initiative 7

John Turner - UCI Data Science Initiative 8

Different viewer types John Turner - UCI Data Science Initiative 9

Different viewer types John Turner - UCI Data Science Initiative 10

Different targeting requirements John Turner - UCI Data Science Initiative 11

John Turner - UCI Data Science Initiative 12 Impression Goals 1 5 2 Impression goals for each advertiser 1

John Turner - UCI Data Science Initiative 13 4 Supply 1 Impression Goals 1 3 5 2 1 2 5 2 We use optimization to find an optimal allocation 1

Combinatorial Explosion Targeting can specify: gender age geographic region income level day of week / time of day the sports you like, etc. Trillions of possible combinations!! John Turner - UCI Data Science Initiative 14

John Turner - UCI Data Science Initiative 15 Aggregate Solve Disaggregate Supply highly variable Expected supply close to zero! Questions: 1. What is best way to aggregate? 2. How to interpret aggregate solution? 3. How good is aggregate solution?

Implications Allow the optimization to segment viewers into buckets This becomes an iterative process Solve à Learn à Revise à Repeat Data requirements change at each iteration John Turner - UCI Data Science Initiative 16

Data- Driven Op'miza'on Sequen&al Process 1. Construct Es&mates of Model Parameters (Forecas&ng / Data Mining) 2. Solve Op&miza&on Model Take-away: Can often get to within 1% of optimality with less than 100 judiciously-chosen audience segments Itera&ve Process 1. Start with Coarse Approxima&ons for Model Parameters (Cheap Forecasts) 2. Solve Op&miza&on Model 3. Learn & Revise Model 4. Re- sample Data to Improve Forecasts 5. Repeat Steps 2-4 17

Example 2: Demand for Trauma Care in Korea l Popula&on ~ 50 million l Area 100,210 km 2 (~ Indiana) l 7 metro ci&es & 9 provinces l Number of trauma cases** = 190,196* l ~ 382.1 trauma / 100k 0 10 50 100 200 400 600 600~ * Stats in year 2008 ** Patients with EMR-ISS score 15 18

Candidate Sites Candidate Trauma Center Sites (n=38) Candidate Heliport Sites (n=54 total: 16 NEMA* + 38 @ TC s) *NEMA = Korean National Emergency Management Agency sites (heliports currently used for fire & other EMS missions) 19

T10H5 T10H15 T10H25 20 Decoupled No-Congestion Integrated Take-away: Solutions of varying quality can look visually similar. For each approach (decoupled, no-congestion, and integrated), indicates the sites chosen for T10H5; indicates the sites chosen for T10H15 but not for T10H5; indicates the sites chosen for T10H25 but not for T10H15

21 Successful Pa'ents (%) Take-away: An integrated model solved approximately is often better than solving two moreprecise models in sequence. 100% 90% 80% 10 trauma centers 12 trauma centers 14 trauma centers 70% 60% 50% Integrated NoCongestion Decoupled T10H5 T10H10 T10H15 T10H20 T10H25 T12H5 T12H10 T12H15 T12H20 T12H25 T14H5 T14H10 T14H15 T14H20 T14H25 # Helicopters # Helicopters # Helicopters

22 Big Data, Big Challenges Conclusions: l Predic&on and Op&miza&on work well together l Predict AND Op&mize, instead of Predict THEN Op&mize.

Data-Driven Optimization John Turner Assistant Professor The Paul Merage School of Business University of California, Irvine October 24, 2014: UCI Data Science Initiative