© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

The Rise of Big Data

October 19, 2017

Jayme Couchene, The Boldt CompanyMark Sands, Performance Building Systems & Catalyst, LLC

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

© LEAN CONSTRUCTION INSTITUTE

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

2

Big What?

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

What is Big Data?

3

• “Big Data” is high -volume, -velocity and -variety information assets that demand cost-effective, innovative forms of information processing for enhanced insight and decision making. – Gartner, Inc.

• “Big Data” refers to the use of predictive analytics, user behavior analytics, or certain other advanced data analytics methods that extract value from data, and seldom to a particular size of data set. – Wikipedia

• Predictive Analytics is the practice of extracting information from existing data sets in order to determine patterns and predict future outcomes and trends.

• “Big Data” is high -volume, -velocity and -variety information assets that demand cost-effective, innovative forms of information processing for enhanced insight and decision making. – Gartner, Inc.

• “Big Data” refers to the use of predictive analytics, user behavior analytics, or certain other advanced data analytics methods that extract value from data, and seldom to a particular size of data set. – Wikipedia

• Predictive Analytics is the practice of extracting information from existing data sets in order to determine patterns and predict future outcomes and trends.

• “Big Data” is high -volume, -velocity and -variety information assets that demand cost-effective, innovative forms of information processing for enhanced insight and decision making. – Gartner, Inc.

• “Big Data” refers to the use of predictive analytics, user behavior analytics, or certain other advanced data analytics methods that extract value from data, and seldom to a particular size of data set. – Wikipedia

• Predictive Analytics is the practice of extracting information from existing data sets in order to determine patterns and predict future outcomes and trends.

• “Big Data” is high -volume, -velocity and -variety information assets that demand cost-effective, innovative forms of information processing for enhanced insight and decision making. – Gartner, Inc.

• “Big Data” refers to the use of predictive analytics, user behavior analytics, or certain other advanced data analytics methods that extract value from data, and seldom to a particular size of data set. – Wikipedia

• Predictive Analytics is the practice of extracting information from existing data sets in order to determine patterns and predict future outcomes and trends.

• “Big Data” is high -volume, -velocity and -variety information assets that demand cost-effective, innovative forms of information processing for enhanced insight and decision making. – Gartner, Inc.

• “Big Data” refers to the use of predictive analytics, user behavior analytics, or certain other advanced data analytics methods that extract value from data, and seldom to a particular size of data set. – Wikipedia

• Predictive Analytics is the practice of extracting information from existing data sets in order to determine patterns and predict future outcomes and trends.

• “Big Data” is high -volume, -velocity and -variety information assets that demand cost-effective, innovative forms of information processing for enhanced insight and decision making. – Gartner, Inc.

• “Big Data” refers to the use of predictive analytics, user behavior analytics, or certain other advanced data analytics methods that extract value from data, and seldom to a particular size of data set. – Wikipedia

• Predictive Analytics is the practice of extracting information from existing data sets in order to determine patterns and predict future outcomes and trends.

• “Big Data” is high -volume, -velocity and -variety information assets that demand cost-effective, innovative forms of information processing for enhanced insight and decision making. – Gartner, Inc.

• “Big Data” refers to the use of predictive analytics, user behavior analytics, or certain other advanced data analytics methods that extract value from data, and seldom to a particular size of data set. – Wikipedia

• Predictive Analytics is the practice of extracting information from existing data sets in order to determine patterns and predict future outcomes and trends.

• “Big Data” is high -volume, -velocity and -variety information assets that demand cost-effective, innovative forms of information processing for enhanced insight and decision making. – Gartner, Inc.

• “Big Data” refers to the use of predictive analytics, user behavior analytics, or certain other advanced data analytics methods that extract value from data, and seldom to a particular size of data set. – Wikipedia

• Predictive Analytics is the practice of extracting information from existing data sets in order to determine patterns and predict future outcomes and trends.

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

It’s All About Data

4

Unfiltered Data

Filtered Data

Big Data

Variation

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Closer Look @ Big Data

5

Variation

Budgeting w/o Functions and KPI’s

Procurement to Delivery

Conceptual Design to Design Development

Functional Program w/KPIs

Final Design & Cost

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Guiding Principle

6

Predictions are only as good as the data supporting them!

Disciplined Data Governance:• Record Data in a Complete, Consistent, & Standard Way• Decipher Critical Data From Junk• Nucleus of Critical Data – Typically Less Than 200 Data Points

Lack of standardization makes comparison extremely challenging

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Which Data to Extract?

7

Start with Critical Data standardized across many organizations

Facility Purpose Departments Functions

Medical OncologyRadiation OncologyPhysician OfficesTherapyAdministration

Linear AcceleratorCT SimulatorPET/CT ScanningHDR ProcedureExams (Specialized)

Cancer Center

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Which Data to Extract?

8

Start with Critical Data standardized across many organizations

KPI’s (Attributes) Key Parameters Schedule

ParkingVertical CirculationCore & Common

Building ConfigurationSite/Building Quantities

Approval to DesignDesign to Mobilize

Mobilize to EnclosureEnclosure to Occupancy

LocationOwner TypeShell Type

QualityDurability

Energy/LEED…and so on

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Systems Approach

9

Start with Critical Data standardized across many organizations

Costs Groups Systems

A – SubstructureB – ShellC – InteriorsD – ServicesE – EquipmentF – Special ConstructionG - Sitework

B – ShellB2010: Exterior WallsB2020: Exterior WindowsB2030: Exterior Doors….and so on

Hard Costs

Soft Costs

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

MacLeamy

10

Ability to Impact Cost Cost of Change

Start FinishTime

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Boldt’s Predictive Analytics Journey

11

Need & Feasibility

Scope & Approval

2017 – Big Data with Systems Approach

Biggest Decisions

Program

Conceptual Design

SchematicDesign

Design Development

Contract Documents

Big Decisions

Bigger DecisionsProcure & Construct

2005 – Systems Approach based on Functional Requirements

Effort Curve Traditional Delivery

Effort Curve IPD Delivery

Effort Curve Big Data

Total Solution

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Extracting Boldt’s Existing Data

12

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Understanding Existing Data

13

Most Efficiently Programmed, Designed and Produced Projects

Data analysis normalized to selected location and time

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Understanding Existing Data - Cost

14

Material and Cost Outliner

Material and Cost Outliner

• 13 of 15 projects produced within 7% of predicted (market average) cost – 2 outliers• Sampling average of all projects is 2.7% less than market average

Sampling Average

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Understanding Existing Data - Program

15

Program and Plan Outliners

Program and Plan Outliners

• 11 of 15 projects exceed 7% variation from of predicted (market average) program and plan• Sampling average of all projects is 4% less than market average

Sampling Average

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Deeper Dive Into the Data

16

Exam Room Illustration

Gross Building Area per Exam Room ranges from:365 SF to 645 SF (~80% difference)

Resulting Cost per Exam Room ranges from:$95,000 to $196,000 (>100% difference)

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Understanding Existing Data Sampling

17

Exterior Wall Illustration

Cost per Wall Area ranges from:$34/SF to $99 (~ 3 times)

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Understanding Existing Data

18

What We Learn From This Data

1. High variation exists at project details (validates studies showing excessive waste)2. Low variation exists at project summary (helps conceal the excessive waste)3. Greatest variation is within facility program/design (Not material specification and building

production.)4. Most projects can be effectively steered to desired outcomes (Starting in the pre-

program stages)5. Project outcomes can be reliably predicted in the conceptual design stages.6. Variations to approved target can be readily discovered and managed.

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Applying This Data to Predict Future Results

19

Conceptual Modeling

Rapid scenario prototyping and dialing in to most likely results – in conceptual modeling stages

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Deming, Lean and Back Again

20

Toyota Production System

Lean Manufacturing

Lean Construction

W. E. Deming

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Big Data and Lean Construction

21

According to Deming

• Process Improvement• Objective Knowledge

• Impartial Prediction and Analysis

• Big Data – Within & Outside The Organization• Standardized Systems Approach

• Disciplined Data Governance

• New Forms of Data Processing

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

Big Data and Lean Construction

22

Impartial Prediction and Analysis

• Target Value Delivery (TVD)

• Choosing By Advantages (CBA)

• Set-Based Planning/Design (aka Concurrent Engineering)

Success Story

© LEAN CONSTRUCTION INSTITUTE

C A P T U R E A N D L E V E R A G E T H E L E A N A D V A N T A G E

In the spirit of continuous improvement, we would like to remind you to complete this session’s survey in the Congress app! We look forward to receiving your feedback. Highest rated presenters will be recognized.

© LEAN CONSTRUCTION INSTITUTE

Jayme CoucheneThe Boldt CompanyProject Development Managerjayme.couchene@boldt.com

Mark SandsPerformance Building SystemsPresidentmsands@buildingcatalyst.com