Multiway Analysis of Bridge Structural Types in the National Bridge Inventory (NBI) A Tensor Decomposition Approach

Similar documents
Lecture 26 Section 8.4. Mon, Oct 13, 2008

Nursing Facilities' Life Safety Standard Survey Results Quarterly Reference Tables

Your Galactic Address

Analyzing Severe Weather Data

Use your text to define the following term. Use the terms to label the figure below. Define the following term.

Sample Statistics 5021 First Midterm Examination with solutions

Evolution Strategies for Optimizing Rectangular Cartograms

Smart Magnets for Smart Product Design: Advanced Topics

Parametric Test. Multiple Linear Regression Spatial Application I: State Homicide Rates Equations taken from Zar, 1984.

Drought Monitoring Capability of the Oklahoma Mesonet. Gary McManus Oklahoma Climatological Survey Oklahoma Mesonet

Appendix 5 Summary of State Trademark Registration Provisions (as of July 2016)

2006 Supplemental Tax Information for JennisonDryden and Strategic Partners Funds

What Lies Beneath: A Sub- National Look at Okun s Law for the United States.

Statistical Mechanics of Money, Income, and Wealth

EXST 7015 Fall 2014 Lab 08: Polynomial Regression

Annual Performance Report: State Assessment Data

SAMPLE AUDIT FORMAT. Pre Audit Notification Letter Draft. Dear Registrant:

Swine Enteric Coronavirus Disease (SECD) Situation Report June 30, 2016

Data Visualization (DSC 530/CIS )

Summary of Natural Hazard Statistics for 2008 in the United States

Final Exam. 1. Definitions: Briefly Define each of the following terms as they relate to the material covered in class.

Forecasting the 2012 Presidential Election from History and the Polls

Class business PS is due Wed. Lecture 20 (QPM 2016) Multivariate Regression November 14, / 44

Module 19: Simple Linear Regression

Cluster Analysis. Part of the Michigan Prosperity Initiative

Data Visualization (CIS 468)

AIR FORCE RESCUE COORDINATION CENTER

C Further Concepts in Statistics

Regression Diagnostics

Swine Enteric Coronavirus Disease (SECD) Situation Report Sept 17, 2015

Meteorology 110. Lab 1. Geography and Map Skills

Discontinuation of Support for Field Chemistry Measurements in the National Atmospheric Deposition Program National Trends Network (NADP/NTN)

Kari Lock. Department of Statistics, Harvard University Joint Work with Andrew Gelman (Columbia University)

Introduction to Mathematical Statistics and Its Applications Richard J. Larsen Morris L. Marx Fifth Edition

The veto as electoral stunt

AFRCC AIR FORCE RESCUE COORDINATION CENTER

Swine Enteric Coronavirus Disease (SECD) Situation Report Mar 5, 2015

Draft Report. Prepared for: Regional Air Quality Council 1445 Market Street, Suite 260 Denver, Colorado Prepared by:

Empirical Application of Panel Data Regression

Effects of Various Uncertainty Sources on Automatic Generation Control Systems

Resources. Amusement Park Physics With a NASA Twist EG GRC

Analysis of the USDA Annual Report (2015) of Animal Usage by Research Facility. July 4th, 2017

APRIL 1999 THE WALL STREET JOURNAL CLASSROOM EDITION Hurricanes, typhoons, coastal storms Earthquake (San Francisco area) Flooding

Introduction to Tensors. 8 May 2014

AC/RC Regional Councils of Colonels & Partnerships

REGRESSION ANALYSIS BY EXAMPLE

Test of Convergence in Agricultural Factor Productivity: A Semiparametric Approach

Outline. Introduction. Introduction Accident Damage. Introduction Act of God Damage NATIONAL HIGH PERFORMANCE CONCRETE FOLLOW UP SURVEY RESULTS BY:

Puerto Rico Transportation Asset Management Plan

Tensor Decompositions and Applications

$3.6 Billion GNMA Servicing Offering

Some concepts are so simple

Combinatorics. Problem: How to count without counting.

Estimating Dynamic Games of Electoral Competition to Evaluate Term Limits in U.S. Gubernatorial Elections: Online Appendix

Blueline Tilefish Assessment Approach: Stock structure, data structure, and modeling approach

Daily Operations Briefing June 9, 2012 As of 8:30 a.m. EDT

Time-Series Trends of Mercury Deposition Network Data

Teachers Curriculum Institute Map Skills Toolkit 411

Clear Roads Overview. AASHTO Committee on Maintenance July 23, 2018 Charlotte, North Carolina

Further Concepts in Statistics

PAVEMENT CRACK CLASSIFICATION BASED ON TENSOR FACTORIZATION. Offei Amanor Adarkwa

Daily Operations Briefing Tuesday, May 14, 2013 As of 8:30 a.m. EDT

The Observational Climate Record

Week 3 Linear Regression I

Topographic Recreational Map Of New Mexico: Detailed Travel Map By GTR Mapping

Section 619. Profile. TA Center. early childhood. national. IDEAs partnerships results. The National Early Childhood Technical Assistance Center

Daily Operations Briefing May 30, 2012 As of 8:30 a.m. EDT

Daily Operations Briefing April 15, 2012 As of 8:30 a.m. EDT

Towards a Bankable Solar Resource

Daily Operations Briefing. August 2, 2012 As of 8:30 a.m. EDT

Further Concepts in Statistics

From Matrix to Tensor. Charles F. Van Loan

Fundamentals of Multilinear Subspace Learning

Estimations of Copper Roof Runoff Rates in the United States

Appendix C. Questionnaire Summary of Responses Geographic Information Systems

A Second Opinion Correlation Coefficient. Rudy A. Gideon and Carol A. Ulsafer. University of Montana, Missoula MT 59812

Daily Operations Briefing February 22, 2013 As of 8:30 a.m. EST

CS-11 Tornado/Hail: To Model or Not To Model

Faloutsos, Tong ICDE, 2009

SUPPLEMENTAL NUTRITION ASSISTANCE PROGRAM QUALITY CONTROL ANNUAL REPORT FISCAL YEAR 2008

2/25/2019. Taking the northern and southern hemispheres together, on average the world s population lives 24 degrees from the equator.

Computational Linear Algebra

Tensor Decompositions for Machine Learning. G. Roeder 1. UBC Machine Learning Reading Group, June University of British Columbia

Transferability of Household Travel Data Across Geographic Areas Using NHTS 2001

DATA UNCERTAINTY IN BRIDGE MANAGEMENT

Daily Disaster Update Sunday, September 04, 2016

Daily Operations Briefing Friday, November 29, :30 a.m. EST

Outline. Administrivia and Introduction Course Structure Syllabus Introduction to Data Mining

Online Appendix for Innovation and Top Income Inequality

2012 USACE Drought Actions

STATIONARITY OF THE TEMPORAL DISTRIBUTION OF RAINFALL Joseph P. Wilson, Wilson Hydro, LLC, Rolla, Missouri

Daily Operations Briefing Tuesday, November 5, :30 a.m. EST

Daily Operations Briefing Friday, January 22, :30 a.m. EST

Drake Petroleum Company, Inc. Accutest Job Number: JB Total number of pages in report:

Risk Assessment of Highway Bridges: A Reliability-based Approach

Simulating the State-by-State Effects of Terrorist Attacks on Three Major U.S. Ports: Applying NIEMO (National Interstate Economic Model)

Locations of Monitoring Stations in the Mercury Trends Network

Daily Operations Briefing. Sunday, March 4, :30 a.m. EST

CP DECOMPOSITION AND ITS APPLICATION IN NOISE REDUCTION AND MULTIPLE SOURCES IDENTIFICATION

Daily Operations Briefing Wednesday, November 6, :30 a.m. EST

Transcription:

Multiway Analysis of Bridge Structural Types in the National Bridge Inventory (NBI) A Tensor Decomposition Approach By Offei A. Adarkwa Nii Attoh-Okine (Ph.D) (IEEE Big Data Conference -10/27/2014) 1

Presentation Outline Introduction Basic Tensor Concepts Analysis & Results Conclusion & Recommendations 2

INTRODUCTION 3

National Bridge Inventory (NBI) Largest database of bridge data in the world 1 Information on >600,000 bridges Owned and operated by the Federal Highway Administration (FHWA) Status of a bridge: not deficient; structurally deficient; and functionally obsolete Basis for determining eligibility for Federal Bridge Funding 4

Structurally Deficient Bridges Structurally deficient bridges have load carrying members in poor condition or significantly below design standards Repair, rehabilitation or replacement required In the US, 1 in 9 bridges is structurally deficient Laid end to end, structurally deficient bridges will span from DC to Denver 2 * http://www.transportationissuesdaily.com/wp-content/uploads/2011/10/sherman-minton-bridge.jpg Sherman-Minton Bridge * 5

Effective use of bridge rehabilitation and replacement funding Need to understand deterioration patterns of bridges under varying loading and environmental conditions NBI can serve as a major resource 6

Focus of Research Analysis of deterioration trends of specific bridge deck structural types in the US Tensor Decomposition 7

BASIC TENSOR CONCEPTS 8

What are Tensors? Generalizations of scalars, vectors and matrices Samples variables Samples Time Time variables variables Samples Weather 9

Multiway data is everywhere Why Multiway Data Analysis? Capture variation in dataset while preserving multidimensional nature of data Reveal hidden patterns in data 10

Typical 3D dataset Subarrays Mode-3 Mode-1 Mode-2 X x 231 Mode-1 Fiber x :21 Mode-2 Fiber x 1:2 Mode-3 Fiber X 11: Horizontal slice X 1:: Lateral slice X :2: Frontal slice X ::3 11

Matricization (Unfolding) Summation: A + B = C, where C iii = a iii + b iii k i χ R i jj i j j j j Outer Product: X = a ο b ο c s.t. x iii = a i b j c k j k Khatri-Rao Product: P I J Q K J = R II J k i χ R j ii j i i i i Kronecker Product: P I J Q K L = R II JJ k i j χ R k ii k k i i i i Hadamard Product: P I J Q I J = R I J j j 12

Tensor Decomposition Reducing data to lower-order forms for analysis (classification, prediction, clustering) 2 main approaches Canonical decomposition/parallel Factors (CANDECOMP/PARAFAC; CP) Tucker decomposition 13

The CP Decomposition Canonical Decomposition/ Parallel Factors a i, b i and c i - factor loadings R- number of components Rank of a tensor=min (R) required to approximate X 14

Choosing the appropriate CP model Alternating Least Squares Approach (ALS) Where is the PARAFAC model % Variance Explained Core Consistency Diagnostic (CORCONDIAG) High CORCONDIAG stable model Low CORCONDIAG invalid and/or problematic model 15

ANALYSIS & RESULTS 16

Where: Structural Deterioration Rate SS i,j,k = n i,j,k N i,j,k j SS i,j,k =structural deterioration rate in state i of structure type j in year k n i,j,k =number of structurally deficient bridges in state i of structure type j in year k N i,j,k =number of bridges in state i of structure type j in year k 17

NBI database (1992-2013) Dataset Structurally deficient bridge design types by state Year State Deck Type Multidimensional Dataset for analysis 18

Selected Bridge Design Types Girder and Floorbeam 4 Slab 5 Stringer/Multibeam 6 Arch Deck 7 Tee Beam 8 Box Beam 9 19

3D visualization 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0 PR WY WI WV WA VA VT UT TX TN SD SC RI PA OR OK OH ND NC NY NM NJ NH NV NE MT MO MS MN MI MD ME MA LA KY KS IA IN IL ID HI GA FL AK AL AR AZ CO CA DE CT DC Girder & Floorbeam system Slab Stringer/Multibeam or Girder Arch Deck Tee Beam Box Beam or Girders(Multiple) 20

Exploration of NBI structure type data slab stringer/multibm or girder girder & flbm Tee Bm Box Beam Arch Deck 0.8 0,4 0.7 0,35 0.6 0,3 0.5 0,25 0.4 0,2 0.3 0,15 0.2 0.1 0,1 0,05 0 Slab Stringer/ Multibeam or girder Girder & Floorbeam Tee Beam Box Beam or girders Box plot of mean SD rates over 22 years Arch-Deck 0 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 SD rate trends for structural types 21

Decomposition plots 2 component PARAFAC model was fitted to the data Variance explained= 90.15% 0.8 5 0.235 Component 2 0.7 0.6 0.5 0.4 0.3 0.2 box bm slab multibm or girder tee bm arch deck gird & floorbm Component 2 4 3 2 1 0-1 -2 ND OK SD AL KY TN LA HI AK CA AZ GAIA AR KS ORNV WAMT UT DC FL MD ID NE MIIL MO PA OH WV VT WI CO MN WY TX NH CT VA PR MA NM RI IN NJ NC ME SC DE NY 0.23 0.225 0.22 0.215 0.21 0.205 0.2 0.195 13 12 11 10 09 08 07 05 06 04 03 02 99 00 01 98 97 96 95 94 92 93 0.1 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Component 1 MS -3-1 0 1 2 3 4 5 6 7 Component 1 0.19 0.17 0.18 0.19 0.2 0.21 0.22 0.23 0.24 0.25 Loading plot for Structure types Loading plot for States Loading plot for years 22

Conclusion & Recommendations Demonstrates tensor decomposition as a knowledge discovery tool that has applications in bridge industry Detailed analysis which incorporates other deterioration factors such as age, ADT, and geographical location. Prediction 23

References (1) Wu N., Chase S. An Exploratory Data Analysis of National Bridge Inventory. Mid-Atlantic Universities Transportation Center (MAUTC). UVA-2009-03. May 2010 (2) Davis S.L., Goldberg D. The Fix We re in For: The State of Our Nation s Bridges 2013. Transportation for America. June 2013 (3) Steel Girder & Floorbeam. http://www.ncdot.gov/_graphics/site_graphics/projects_ncbridges_hb_t_steelgirder01.jpg. Accessed:10/21/2014 (4) Slab bridges. http://www.ncdot.gov/_graphics/site_graphics/projects_ncbridges_hb_t_conslab01.jpg. Accessed:10/21/2014 (5) Steel stringer or multibeam. http://www.ncdot.gov/projects/ncbridges/historic/types/?p=13#types. Accessed:10/21/2014 (6) Arch Deck. http://www.asce.org/cemagazine/article.aspx?id=25769811288#.vebr8pldwl0. Accessed:10/21/2014 (7) Tee Beam. http://www.ncdot.gov/_graphics/site_graphics/projects_ncbridges_hb_t_teebeam01.jpg. Accessed:10/21/2014 (8) Box Beam bridge. https://www.michigan.gov/documents/mdot_rr-102_143535_7.pdf. Accessed:10/21/2014 24