Data Fabric for Departments of Transportation: An Enterprise Approach
/What Every GIS Professional Needs to Know
Our introduction to data fabric blog explains data fabric, how to evaluate an organization’s data maturity, and data fabric architecture. Departments of transportation (DOTs) face the challenge of managing, integrating, and leveraging agency information.
A key point to remember is that geographic information systems (GIS) are data integrators. Most people don’t realize that the central hub of a GIS environment is bringing data together and making it available for visualization, not just spatially, but also through reports and dashboards.
What is a Linear Referencing System (LRS)?
There are multiple definitions for a linear referencing system. Considering the data fabric context, our definition is: a set of procedures and methods for specifying a location as a distance, or offset, along a linear feature from a point with a known location.
An LRS does not store data; it allows you to locate data along the network and keep it stored in the appropriate business or data system.
An LRS is a set of procedures and methods for specifying a location as a distance along a linear feature from a point with a known location.
A common misconception is that an LRS supports only one department of an organization. In reality, an LRS supports all business partners.
Linear Referencing Methods
For an LRS to work, you need a linear referencing method (LRM). An LRM allows you to locate the data and measure the distance from a certain point. DOTs commonly use LRMs to describe events relative to each other. Common examples include:
Highway milepost markers known as reference posts: Post 114 and offset .2 miles on I-35.
Mile point: 34.68 miles from the start of route I-35.
Coordinate and route combination: 42.017901, -93.571195, and route I-35.
Physical addresses: 207 Senate Ave., Camp Hill, PA 17011-2316.
Literal description: 200 feet east of the intersection of Maple Ave. and Main St. on Main St.
LRS Transformation
Think of LRS transformation as Google Translate for DOTs. One reason to have an LRS is to transform data from one LRM to another. LRS data stored in different LRMs requires many transformations for the data to be useful. Traditionally, organizations write their own transformations, which is expensive and inefficient.
Understanding Dynamic Segmentation
Dynamic segmentation groups data based on specific criteria that can change over time. A step-by-step dynamic segmentation process is as follows:
- Use your common route geometry from the LRS.
- Break up the route geometry by attributes into independent geometries (dynamic segments).
- Speed limit.
- Pavement conditions.
- Number of lanes.
- Traffic volumes.
- Store these breaks in your LRS or business systems.
A linear overlay allows you to take the dynamic segments and overlay each segment to create segmentation-based outputs to understand patterns and identify trends.
Linear overlays combine individual components – i.e., annual average daily traffic (AADT), number of lanes, speed limit, and pavement conditions – and result in pattern identification.
Dynamic Segmentation for Transportation Agencies
There are multiple examples of how DOTs use dynamic segmentation. The following example highlights three different types of transportation data collection, each using a different linear referencing method. The LRC can transform the location between different LRMs.
Dynamic segmentation allows different types of transportation-related data to be collected along the same route for different purposes.
Dynamic Segmentation and Traffic Incident Management Plans
The Iowa DOT uses dynamic segmentation to manage its traffic incident management (TIM) plans by dynamically segmenting its network into sections. They can leverage web services to access real-time data for traffic information, maintenance, etc. Click here to access the complete document.
These traffic management segments can be overlayed with other linear data for advanced analyses.
Dynamic segmentation’s applicability in traffic incident management allows different departments and agencies access to the same data.
How a Department of Transportation Uses Data
Our introduction to data fabric blog identifies the data fabric architecture and explains how it provides a foundation for the entire data lifecycle.
The Power of Open Data
There are two basic types of open data:
Open and available within an organization.
Open and available to the public.
Data fabric makes data open and available through web-based services. Leveraging GIS to establish data governance, an Esri® hub portal assists with making data available to other departments or public agencies. DOTs can leverage Esri open data tools and service environments.
Benefits of Using Open Data
There are many benefits to using open data, including:
Leveraging GIS to integrate regional data into state data, resulting in more informed decisions.
Promoting cooperation across agencies and within departments.
Establishing and enforcing data standards for easier data sharing.
Creating relationships with the public and road users throughout the data lifecycle.
Data Visualization
DOTs make strategic investments in many software platforms. To receive the most significant return on investment, it is a best practice to allow – and encourage – employees to use the software programs to their fullest extent. Many programs have out-of-the-box functionality, including Esri dashboards.
Master Data Management
Master data management is the heart of data fabric. It brings the processes, strategies, and technologies together across a DOT, ensuring accuracy, consistency, reliability, and authority. A master data management ecosystem consists of the following:
Source data (collecting internal and external data).
Data staging (processing).
Data warehouse (storing).
Information systems (analyzing and editing).
Case Study: Machine Learning and the Iowa Department of Transportation
A real-world example is a project I worked on at the Iowa Department of Transportation. We equipped snowplows with cameras to better visualize the road conditions by DOT staff and the public. We took the image data and used machine learning to determine if the pictures could report the road conditions automatically. After training the machine with more than 25,000 images, the snowplow images were 90% accurate.
Unlock the Power of GIS
Reject the theory that GIS is a standalone technology. GIS is an integrating technology.
Almost everything in a DOT has some location component: road conditions, traffic signals, asset management, striping, etc. With a location at the core of transportation data, spatial analysis and visualization are the next logical step in the data lifecycle. GIS plays a critical role in enterprisewide data integration.
Data Standards
Having data standards is imperative to maintain authoritative data. GIS can help build and enforce standards, especially regarding data organization and storage. Some DOTs can get stuck in analysis paralysis. One way to avoid this is to identify the critical data sources that need to be integrated and start there.
Data Collection
An intelligent transportation system (ITS) is one example of DOT data collection. These systems collect vast amounts of data that can be used to inform many decisions and future planning efforts, including traffic management and control, electronic toll collection, parking management, and emergency management.
Applying Data Fabric for DOTs
DOTs face various data management challenges, which will continue escalating exponentially. Embracing the data fabric approach and combining it with the power of GIS results in integrated data and unified data environments. With data fabric as the foundation, DOTs can continue their digital transformation by harnessing the power of their data.
About the Author
Eric Abrams
Senior Project Manager
Email Eric
Connect on LinkedIn
Get to Know Eric
