Spatial Analysis Helps Utilities Manage Risk

Creating a better risk model

When dealing with the complex infrastructure of an electric, gas or water utility system, things often go wrong. Things go wrong because there are so many factors that can contribute to a problem. Utility operators face an enormous task. They must gather accurate and timely data, understand the relative importance of each factor, and determine relative risk of damage to the system. Once utility risk is understood, a rational mitigation and investment strategy can be developed. Most utilities are able to prioritize maintenance and replacement projects based on factors such as equipment age, and the history of maintenance, operation, and failure.

Despite such measures, unexpected things happen. In the event of an outage or leak, you will often hear experienced field workers say, “I knew this water main would burst,” or “I suspected this transformer would fail,” and “That gas main has always been troublesome.”

The problem is this: Some of the factors that often contribute to system failure or add risk of failure are not systematically built into the utility risk model. What these experienced workers are doing is a form of spatial analysis in their heads.

Experienced workers know that a pole at the bend of the road is more apt to be hit by a car than a pole along a straight-a-away. A transmission line that crosses a river or canal has a higher risk of being damaged due to river-way traffic. A direct buried cable is more likely to fail if it has experienced a deep freeze followed by a fast thaw if it is buried in rocky soil. While crucial infrastructure information is often known by employees, it is hard to quantify in a risk model. The key to better risk management is a risk model with spatial analysis that reaches both within and outside the utility.

Utilities can now access all kinds of information online and easily incorporate it into a risk model using GIS. We must look to web-based data sources, and take advantage of geo-enabled handheld devices to help build a better risk profile. In addition to internal data, utilities have access to a wide variety of information related to weather, soil, flood patterns, hazards, newsfeeds, and more. Sources include predictive and measured data as well as social media data.

Further, a utility can update its risk model to include information from those experienced workers who have a qualitative understanding of the company’s infrastructure vulnerabilities. Most of this information is spatial in nature, and can be collectively organized on a GIS platform for risk analysis. GIS has the most convenient way of presenting the results of the analysis–in the form of an interactive map that can be viewed over the web, in the field, and from the desktop.

How can GIS technologies and spatial analysis be more readily employed by utilities to enhance their risk models?

Bill Meehan

About Bill Meehan

Bill Meehan, P.E. heads the worldwide utility practice for Esri. Author of Empowering Electric and Gas Utilities, Power System Analysis by Digital Computer, and numerous papers and articles, Bill has lectured extensively and taught courses at Northeastern University and the University of Massachusetts. Follow Bill on Twitter.
This entry was posted in Industry Focus and tagged , , . Bookmark the permalink.

Leave a Reply

4 Comments

  1. Bob Montgomery says:

    Combining spatial analysis with traditional geospatial systems can create a new decision support system for use in operations decision making, risk management analysis, and business case development. Spatial analysis allows engineers and decision makers to mashup authoritative data, predictive data, measured data, and inputs from subject matter experts and social media. All of this means that data existing within the utility IT systems (such as SCADA, GIS and OMS) can now be combined with data from the cloud.

    As mentioned in Bill’s post, cloud data could include data such as weather, lightning strikes, soils, crime statistics, and consumer photos from iPhones or tweets. Additionally, spatial analysis could include knowledge transferred from veteran utility employees into a data model. This solution could empower decision making in areas such as outage restoration, economic development, engineering, and financial analysis. The results of these analyses could then be distributed using devices such as iPhones and iPads, without requiring special licensing or additional software.

    Spatial analysis, combined with geospatial systems, can incorporate risk management into every decision and result in better solutions being implemented. The tipping point has now been reached with spatial analysis tools and cloud computing. Utilities have a great opportunity to embrace spatial analysis to protect their investment in geospatial systems, minimize risk, and maximize value in daily operations.

  2. Nice post, Bill. This reminds me of another round table discussion we had here a while ago — Capturing institutional knowledge of experienced field crew. Utility data model should be revamped to incorporate such knowledge. Social media models can be borrowed to collect institutional data. From a simple ‘comment’ tool to geo-tagged real-time hashtags (tweets) shared among field workers during field data collection could prove practically beneficial for better utility planning and risk management. GIS, now, facilitates such technologies through social media maps (APIs). Coupling with semantic web is another idea yet a long way to go.

  3. Bill Meehan Bill Meehan says:

    Thanks for the post. The beauty of spatial analysis is that it can use a variety of source material. I like to think of the data that goes into the spatial analysis as falling into one of five categories, authoritative (stuff you know to be fact, like age), predicative (things that are likely to happen – like flooding prediction maps), measured (from SCADA and weather stations), experiential (from field workers) and community (from crowd sourcing, social networking, tweets, etc). Add this all up, weight it appropriately and you get a much clearer picture of risk.

  4. Cindi Salas says:

    While I can’t take the credit for this, I recall a paper I read recently that discussed geospatial intelligence as actionable knowledge. This paper noted:

    • It is the ability to perceive, understand, interpret, and operate to anticipate the impact of an event or action within a spatiotemporal decision-making environment.

    • It is also the ability to collect, store, and manipulate geospatial data of the present to create knowledge through critical thinking, geospatial reasoning, and analytical techniques.

    • It is the ability to present the possible futures in a way that is appropriate to the decision-making environment and provides new insights.

    (Bacastow, T.S. and Bellafiore, D.J. (2009). Redefining geospatial intelligence. American Intelligence Journal. Pp 38-40)

    Since utilities were early adopters of GIS for maintaining records of their infrastructure, we simply need to the leverage this tremendous investment—this treasure trove of asset-rich data that in many cases already exists. And we need to do so thinking beyond the normal uses of just knowing where the assets are and how to navigate to them efficiently.

    For instance, at CenterPoint Energy, we are currently utilizing our smart meter infrastructure information (meters, cell relays, take-out points) and incorporating current and historical periodic failed interval readings to a do predictive analysis of more potentially serious system problems.

    Other examples include utilizing historic leak repairs to predict gas lines with the highest probability of failure, as well as analyzing loaded poles relative to various variables such as wind and ground saturation that could result in their failure.

    I believe, to a large extent, we have a lot of what we already need in place. But we need to challenge ourselves to put this information to work. The possibilities of spatial analysis for risk mitigation are numerable, and the value is priceless.