April 2019
THE DATA CENTRIC TOLLING ENTERPRISE
Building a Roadmap for Success
April 2019
April 2019
PRESENTERS
2
DAN MONTGOMERYNorth Highland Data & Analytics
Information Management Lead
972-523-5367
April 2019
AGENDA
3
Background / Challenge
Objectives & Expected Benefits
Approach
Current State Insights
Future State
Strategy & Roadmap
1
2
3
4
5
6
Questions7
April 2019
BACKGROUND / CHALLENGE
4
TxDOT Tolling Overview
- Vision: "A customer-focused service provider that creates economic opportunities, stimulates investment
and enhances quality of life by supporting high-performing transportation systems in Texas and beyond."
- Mission: "Exceed customers' expectations and provide leading mobility solutions by delivering and
growing an integrated, safe, reliable and efficient highway system."
- 440M annual transactions
- $410M annual toll revenue collected
- 275 Toll Road Miles
- Toll roads in the Austin, Dallas, and Houston metros
- TxDOT Tolling was looking to make strategic decisions by leveraging data coming from three primary
sources as well as other sources from multiple origins
April 2019
OBJECTIVES & EXPECTED BENEFITS
5
The following are the key objectives and anticipated benefits of becoming a Data Centric Organization
KEY OBJECTIVESCreate a sustainable and actionable Data Warehouse and Business Intelligence strategy
Develop a best practice Data Governance framework to support the operationalization of the DW/BI roadmap
KEY BENEFITS• Promotes fact-based decision making through the effective utilization of data assets
• Creates a common language for critical information
• Aligns business strategy with information
• Enhances data trust
• Standardizes processes
April 2019
APPROACH - DATA COMPONENT MODEL
6
The data component model consists of 12 components that align your information to the business strategy through
people, process, and technology
GOVERNANCE• People (governance organization including data stewards)
• Process (best practices and procedures)
• Technology (tools to support data governance)
OPERATIONAL COMPONENTS• Master Data Management
• Data Quality
• Metadata Management
• Analytics
• Dashboards, Scorecards, and Reporting
• Security and Privacy
• Data Integration
• Data Strategy and Architecture
PROJECT EXECUTION COMPONENTS
• Project Management
• Change Management
• Organizational Alignment
BUSINESS
STRATEGY
Master Data
Management
Data Strategy
& Architecture
Data
Quality
Metadata
Management
AnalyticsDashboards
Scorecards
Reporting
Security &
Privacy
Data
Integration
April 2019
APPROACH - THE PHASES OF THE DW/DG/BI SOLUTION
7
Where you are today Where you want to be How you get there
Siloed, manual data operations
limiting the ability to fully
leverage data
Data-centric operations
promoting analytical insight and
enabling the realization of
strategic initiatives
Advance data competency
through the execution of key
BI/DW/DG activities
Together, we set out to accomplish the key objectives through the execution of three phases
DISCOVERY FUTURE STATE ROADMAP
April 2019
CURRENT STATE INSIGHTS
8
DATA AS AN
ASSET
SILOED
OPERATIONS
METADATA
MANAGEMENT
DATA
CONTINUITY
DATA
INTEGRATION
DATA
GOVERNANCE
Limited view of the role
of data as an asset
(transactional output vs
data as a driver of
analytics)
Siloed functional analysis
creates operational
inefficiencies
Reliance on function-
specific knowledge vs.
common metadata
definitions across the
division
Reconciliation between
source systems requires
extensive manual effort
to maintain data
continuity, further limited
by vendor contracts
Decentralized BI
capability reduces the
ability to integrate data
sources for robust
analytics
Lack of data governance
hinders progress in
building an integrated
environment for BI and
analytics
Significant effort spent
on data reconciliation
vs. data analysis
Duplication of effort
due to lack of
knowledge sharing and
organizational
alignment
Lack of cross-
functional knowledge
management, risk with
employee retention,
implications for
customer sentiment
Potential misalignment
across reporting output
requiring additional
reconciliation; access
limitations due to
vendor reliance
Limited capability to
derive action-oriented
insight and effectively
manage public
relations / bond
reporting
Inhibiting ability to take
data competency to the
next level
Several common themes on the current state of tolling data operations, with the following demonstrating some examples.
Tolling agencies can expand operations (through interoperability) and achieve strategic initiatives with the current state of data operations
TH
EM
EIM
PLIC
ATIO
N
1 2 3 4 5 6
April 2019
FUTURE STATE
9
Siloed, manual data operations limiting
the ability to fully leverage data
HOW YOU GET THERE
TODAY FUTURE
STRATEGIC INITIATIVES
1. Increase TxTag participation rate
2. Replace back-office system and
operator
3. Become nationally interoperable with all
toll facilities in the U.S. and Mexico
4. Build a data warehouse to store and
manage records
5. Build TxDOT Toll Operations Division
staff to operate with the knowledge to
manage a world class customer service
organization that is both customer
focused and efficient.
CURRENT STATE OF DATA OPERATIONS
• Limited view of the role of data as an
asset (transactional output vs data as a
driver of analytics)
• Siloed functional analysis creates
operational inefficiencies
• Reliance on function-specific
knowledge vs. common metadata
definitions across the division
• Reconciliation between source systems
requires extensive manual effort to
maintain data continuity, further limited
by vendor contracts
• Decentralized BI capability reduces the
ability to integrate data sources for
robust analytics
• Lack of data governance hinders
progress in building an integrated
environment for BI and analytics
Data-centric operations promoting
analytical insight and enabling the
realization of strategic initiatives
Advance maturity across the data component
model through the execution of key BI/DW/DG
initiatives and projects
Focus of today’s discussion
April 2019
FUTURE STATE – DW/DG/BI BICC & DATA GOVERNANCE
10
The Business Intelligence Competency Center (BICC) and Functional Data Governance Organization were selected during the Future
State Visioning Session as the core pillars to the Future State Organization to drive the advancement of data competency
BUSINESS INTELLIGENCE
COMPETENCY CENTER
The BICC is comprised of a
centralized team closely aligned
with functional partners across
the division
FUNCTIONAL DATA
GOVERNANCE
The data governance organization
is built around the functional area
requirements and usage of data
April 2019
FUTURE STATE - ORGANIZATIONAL MODEL
11
The data competency organization integrates the BICC and DG structure to operationalize the future state
BI Analyst
TxDOT Toll Operations Division
TOLL EQUIPMENT
& FACILITIES
REVENUE &
BUDGET
OPERATIONS
BOS
ENGINEER &
PLANNING
COMMUNICATION
& MARKETING
PMO (Project)
BUSINESS
SERVICESBICC
BI Analyst
To Be Discussed
Data Governance Steering Committee
Data Governance Manager / BICC Leader
DC
Data
Steward
Data
Steward
Data
Steward
Data
Steward
Data
Steward
Data
Steward
Data
Steward
DC = Data Custodian
DC DC DC DC DC DC DC DC DC DC DC DC DC DC
Note: BI Analyst and Data Steward roles to potentially be fulfilled by the same named resource
BI AnalystBI AnalystBI Analyst BI Analyst BI Analyst
BICC Leader, Pgm. Mgr.,
BI Bus. and Tech Arch.,
Data Arch., Report and
ETL Developer, Bus.
Analyst, DBA, QA/Tester,
Chg. Mgmt., Metadata
Specialist
BI Analyst
*individuals on the transition team
April 2019
FUTURE STATE - REFERENCE ARCHITECTURE
12
Tableau
Back Office Vendor
BI Tool
Data Discovery
Tools
TBD - Advanced
Analytics Platform
TOAD / Other
DATA STORAGE
Data
INF
O P
OR
TA
L
AN
AL
YTIC
S W
OR
KB
EN
CH
DA
TA
SC
IEN
CE
LA
BO
RA
TO
RY
ANALYTICS ACCESS
Customer
HubMetadata
Store
Event Data?
CURATED- Conformed
dimensions
- Subject areas
BUSINESS
ACCESS- Data Marts
- ROLAP
SOURCE TYPES DATA
MOVEMENT
DATA WAREHOUSE
Archive Data
MASTER DATA
MANAGEMENT- Customer Data Hub
- Repeatable process
DATA QUALITY
MANAGEMENT- Source data quality
- KPI validation & audit
SECURITY & PRIVACY
CONTROL- Data classification
- Tokenization management
- Access control/monitoring
Processes
and Controls
ARCHITECTURAL
STANDARDS- Data/Platform/ETL
standards
DATA
GOVERNANCE- Data ownership
- Data stewardship
- Data lifecycle management
METADATA
MANAGEMENT- Data lineage
- Data dictionaries
RAW- Landing
- Validation
(optional)
Marketing
Social Media
Twitter, FB
T&R Data
TxTAG
Web Site
Weather
Events
TBD
Back Office
New
Cust. Svc.
Conduent
Road Usage
Lonestar
Crash Data
CRIS
Vehicle
DMV
Back Office
Conduent
Expense
Peoplesoft
Lane Data
TransCore
Informatica
Back Office Vendor ETL
Tool
Talend
April 2019
STRATEGY – FOCUS ON BUSINESS “ANALYTICS” NEEDS
13
CA
TE
GO
RY
TOLLING OPERATIONS
Understanding operational data, both from
the user’s and lane’s perspectives
TRAFFIC PATTERNS
Discerning trends in traffic
and the factors that
influence them
CUSTOMER INSIGHT
Describing customers and
measuring how well they
are being served
FINANCIAL PERFORM.
Providing insight into the
agency’s past, current, and
future financial health
SU
BC
ATE
GO
RY Tag
Using information
sent from driver’s
TxTags
LaneGathering
Information from
lane equipment
Interop.Collecting tag data
used in
interoperable
settings
RegularIdentifying typical
traffic patterns that
are seen on tolls
apart from
idiosyncratic events
Event/WeatherTracking and
predicting how
idiosyncratic events
affect traffic
patterns
BehaviorMeasuring driver
qualities based on
certain actions and
characteristics
ServiceDetermining the
level of service that
is offered to
customers
RevenueCalculating
received and
anticipated
payment and
funding
ExpenseComputing the
costs incurred and
is expected to be
incurred
US
E C
AS
ES
1. TxTag stats
2. iToll Analysis
3. TxTags on Home
4. Tag Usage
1.Daily Monitoring
2. Monthly
Monitoring
3. Speeds GP vs ML
4. Trans. Type
5. Axle Class
6. Postings
1. Away Tag
Analysis
2. TxTags on Away
1. Traffic Control/
Management
2. Value of Time
3. Safety
4. Predictive
Equipment
Maintenance
1. Timeline/
Duration
2. Lane Variation
3. Managed Lanes
Predictive Analysis
1. Type of Accounts
2. Zip Code Analysis
3. Trend and
Predictive Cust.
Behavior
4. Habitual Violator
5. Trips
1. Customer
Account Analysis
2. Fraud
3. Cust. Service
Stats Analysis
4. Call times
5. Website
Response Time
6. Page Refresh
7. Click to Pay
1. Revenue
Recognition
2. Rejected
Transactions
3. Collections
4. Aging Analysis
5. Invoice Stats
6. Lane to Back
Office
1. Budget &
Expense Monitoring
2. Contract &
Encumbrance
Information
3. TCC Metrics
4. TSA
Each data use case has been grouped with similar use cases, which will be developed during the Phased Builds
1 2 3 4
Note: In addition, dashboard views for reporting to various stakeholders – including legislature, districts, bondholders, and drivers –
may be compiled on an as needed basis using the above use cases.
April 2019
STRATEGY – QUICK WINS AND OTHER KEY ACTIVITIES
14
QUICK
WINS
FOUNDATIONAL
ACTIVITIES
PHASED
DW BUILD
Objective Maximize immediate value of the
FS DW / DG / BI solution
Standup DW / DG / BI infrastructure
for sustainable operations
Accelerate operationalization of the
DW / BI solution
Example of Key
Activities
• Create centralized event tracker
• Define terms of reference
• Standup DG steering committee
• Conduct organizational alignment assessment and present key recommendations
• Formalize the business strategy with a detailed execution plan to ensure cross-functional alignment
• Develop communication plan to begin socializing the value/benefits of the DW / DG / BI solution
• Build source system data visualization to demonstrate immediate value
• Confirm technology platform for the data warehouse (long-term viability, short-term solution)
• Select/confirm appropriate required technologies/tools
• Detail high level DW data architecture
• Define metrics / KPIs across functions
• Establish BICC
• Setup DG program (People, Process, Technology)
• Collect District / Region requirements for DW / BI
• Determine interoperability requirements
• Define metadata solution / establish data literacy
• Develop Master Data Management (MDM) Program
• Execute Phased build of DW / BI (multiple iterations to deliver business case capabilities)
• Each phase/iteration is intended to deliver one to many business case capabilities/functionality. We will be grouping the “business case capabilities/functionality” into phase/iterations as part of the roadmap phase
• The average timeframe for each phase/iterations is approx. 90 days but could vary from 60 to 120 days
1 2 3
The key activities around future state realization have been grouped into three key categories
April 2019
ROADMAP - DATA USE CASE PRIORITIZATION
15
Use cases will be identified and prioritized based on those use cases with the perceived highest value and are most
feasible to implement
FEASIBILITY CRITERIA –
• # of source systems
• Quality of the data
• Ease of source data access
• Volume of historical data required to be loaded
• Number of data transformations required
BUSINESS VALUE CRITERIA –
• Tied to Strategic initiatives
• # of business functions impacted
• Efficiency gains
• Improved quality / usability
• Actionable output (dashboard view vs. detail
spreadsheet view)
LOW HIGH
HIG
H
Business Value
Imp
lem
en
tati
on
Fe
asib
ilit
y
HIGH PRIORITY*
Use Case 1
Use Case 2
Use Case 5Use Case 4
Use Case 3
*DW/BI development groupings to be determined based on Value/Feasibility plus other criteria, such as common source systems and Division priority.
April 2019
ROADMAP – ACTIVITY EXAMPLES
16
ACTIVITY 1 2 3 4 5 6 7 8 9 10 11 12 13
Stand up DG Steering Committee
Formalize the business strategy
Conduct organizational alignment assessment
Develop communication/change management plan
Set up DG program
Define metrics / KPIs across functions
Create centralized event tracker
Define first 50 terms of reference
Develop source system data visualization
Confirm technology platform for the data warehouse
Establish BICC
Select/confirm required technologies/tools
Collect District / Region requirements for DW / BI
Determine interoperability requirements
Detail high level DW data architecture
Define metadata solution / establish data literacy
Develop Master Data Management (MDM) Program
Execute phased build of DW/BI*
Months
*Phased build to include multiple iterations, informed by data use case prioritization
April 2019
QUESTIONS?
THANK YOU!
Building a Roadmap for Success
April 2019
Top Related