National Grid Webinar: Answering Your Questions


Recently David Salisbury, Head of Network Engineering for National Grid and Neil Barry, Senior Director EMEA at Space-Time Insight, presented the webinar “How Analytics Helps National Grid Make Better Decisions to Manage an Aging Network“, hosted by Engerati.  [Listen to the recording here.] Unfortunately, not all the submitted questions were able to be answered in the time allotted.  However, responses have been provided in this post.

How were pdf data sources incorporated into your analytics? How will that be kept up to date?

To correct to the discussion in the webinar, pdf data sources were not analysed in the valves and pipeline use cases. For the corrosion use case, data from pdf reports was manually rekeyed into the analytics solution.


Are there mechanisms built into the system that facilitate data verification and data quality monitoring?

In the general case, metrics were computed for data completeness (e.g., of the desired data, how much was actually available) and confidence (e.g., how recent was the data we used). For the corrosion use case, there are checks for data consistency and completeness.  For pipelines and valves, these metrics have not yet been fully configured.


Could you describe how this helps with the audit trail?  As the system changes, the current snapshot is updated.  How do you show the status at a certain point in the past when a decision was made?

For the corrosion use case, the history is stored and accessible, providing an audit trail. The foundation analytics does offer a ‘time slider’ that delivers animated time series data, making it easy for the user to go back in time.  However, this is not currently configured for National Grid.


Please provide specific examples of how decisions were made based on analytics and demonstration of analytics/predictive analysis

David described an example at around the eight minute mark into the webinar – budgets used to be set locally, but the insight from analytics might show that a particular type of problem is located in a specific geographic area. This can help with decisions around investment and risk.


How have you defined Asset Health? What data is required to assess?

Models for asset health were agreed upon by National Grid and Space-Time Insight during the implementation process. For pipelines, as was mentioned in the webinar, two of the data sets are Close Interval Potential Survey (CIPS) and Inline Inspection (ILI). For valves, a number of data sets are used, including test results and work orders.


Did you look at techniques to predict issues based on historical data…so you can target risk areas?

This has not been implemented by National Grid.  However, the product software has the capability to predict the probability of failure and the criticality of that failure, as one example.


Has Space Time insight worked on developing a situational intelligence tool for electric distribution and/or transmission applications? Similar to the gas transmission monitoring developed for National Grid?

Yes, Space-Time Insight offers an asset intelligence solution for electricity transmission and distribution utilities.  More information is available online.


UK Smart Meter Roll Out: It’s All About The Data



Despite some 2.6 million smart meters already being installed in the UK, it is the data infrastructure that is causing delays with the further roll out of smart meters in the UK, according to a recent BBC article. This IT project is necessary to support the volume of data anticipated to come from the smart meter roll out that is being pushed by the government.

From the chart below you can see how many meters have been installed since 2012.  Higher volumes of data are already being collected which reinforces the need for this important IT project to be up and running as soon as possible.


(Chart and data available from the UK Department of Energy & Climate Change)

With news that the data infrastructure launch is pushed back until the autumn, what impact will this have?

How much data will smart meters generate?

To do a quick calculation on monthly meter reads from the potential smart meters across the UK, there would be around 53 million reads per month. By contrast, with smart meters that record data every 15 minutes, this could mean 96 reads a day from 53 million meters resulting in thousands of times more data being generated. This is obviously a rough estimation but gives an indication as to what the energy companies would be dealing with. This doesn’t include status messages from the meters which would add to the mass amount of data being generated.

Why is this so important, if smart meters are just about making billing automated and putting an end to manual meter reading? There is a lot more value within meter readings and status messages beyond billing.

The benefits of smart meters are clear for consumers: tracking how much energy you are using, monitoring the effect of changes that you have made to your energy consumption, and receiving accurate bills without having to submit a meter reading.

When applied properly, data helps energy companies to manage supply and demand in a much easier fashion. Energy companies benefit from analysing the data collected from the smart meters to enable new rates and business models, implement demand response programs, manage solar power panels in a better way and improve support for electric vehicles, to mention but a few.

To benefit from the thousands-fold growth in meter data, energy companies need analytics that locate the problems and opportunities hidden inside this massive amount of data. Smart meter analytics must be intelligent enough to do the heavy lifting for users, not just make it easier somehow for users to browse among millions of meters. Increasingly, analytics for this size of data set needs the intelligence and autonomy to make decisions independently.

Once the IT infrastructure is in the place, the UK energy companies can start pursuing the new value within smart meter data, analysing it to make better business decisions. All 53 million UK meters likely won’t be changed out by 2020, but that shouldn’t stop UK energy providers from using the smart meter data they already have, or will have soon.

(Image courtesy of rido / 123RF Stock Photo)


Analytics Makes Water Conservation More Precise—and Fun!

water sprinkler
Watering sidewalks wastes water.

While human populations shift, migrate and grow, the amount of fresh water remains constant. (The only way to make large quantities of fresh water is desalination, which is energy-intensive and practical only near the ocean.) To serve more people from the same supply of water, communities increasingly are turning to conservation measures.

In times of drought, communities employ water usage restrictions such as banning car washing and limiting lawn watering to certain days of the week or month. These rules rely on observable behaviors that aren’t easily enforced unless the utility employs smart meters or water cops—or both.

Analytics provide water utilities with unique ways to encourage participation in conservation without smart meters and water cops. Here are a couple of examples.

Pinpoint water conservation opportunities

Some utilities offer financial incentives such as loans or rebates for people willing to take the conservation steps of removing their lawn. According to the U.S. Environmental Protection Agency,  landscape irrigation nationwide accounts for nearly one-third of all residential water use, totaling nearly 9 billion gallons per day.  Which customers are likely to participate?

A first step for utilities is identifying the areas where such a lawn removal program can be successful. Geo-spatial analytics identify portions of the utility service territory where home footprint is small relative to lot size, which shows customers with a potentially high percentage of irrigable land. Locating such lots in areas with less tree cover shows yards that may require more watering in hot months. Sorting a list of these houses by the size of their seasonally-adjusted water bill shows you customers who might be motivated to lower their water bills by replacing some grass with hardy, native perennial shrubs and ground covers.

You might want to filter out houses with pools and hot tubs, to prevent those large water uses from skewing your results. Also, you could focus on households without children living at home. These households might lack built-in labor for mowing and be paying a lawn care company as well as high water bills.

Make conservation a game

Gamification applies concepts such as skills, challenges, points and rankings to non-game contexts such as conservation. Gamification brings out the competitor in us and can make conservation if not fun, then at least a little more rewarding and less boring.

One simple conservation game simply shows you how your water use compares with houses geographically close to you and with households that resemble yours in terms of lot size, people in household, and number of bathrooms. Improving your standing relative to those like you earns you points and entitles you to prizes.

Another, more active game may be to progress through levels of conservation skill, knowledge and savings by accepting certain challenges. Challenges could include doing a home water audit using the utility’s online tools, installing free low-flow showerheads from the utility, or cashing in rebates on low-flow toilets. You receive points as you complete challenges and lower your consumption. Analytics show you how you compare with your neighbors who are and aren’t playing the game. Achieving certain levels in the game might entitle you to prizes such as gift cards or bill rebates.

Fairly simple analytics run behind the scenes in these games to award points and keep score. More sophisticated analytics might be used to recruit players for the game and to target the challenges and rewards to more closely match the water saving needs and opportunities of their neighborhood and household type.

(Image courtesy of ewapee / 123RF Stock Photo)


Augmented Reality For The Enterprise: A Use Case In Electrical Substation Field Service



Augmented reality can make a real impact in field service workers in almost any industry. For a specific example, let’s look at a field technician visiting a problematic transformer at a substation.

Currently field technicians download test plans and view them on a laptop or iPad. They have to remain near their car as they conduct spatial reasoning with regards to electrical circuits and their locations. They have to hold some of they learned and deduced in memory, meaning they are eating up more bandwidth of the part of their brain that holds memory, conducts planning and spatial reasoning or navigation. This forces them to continually reference back and forth between the computer screen, supporting documents, tools and the work site. Many of the artifacts they will use are situated at angles and distances where they must turn away and effectively interrupt the mental process of fixing their eyes on the object they plan to use to do their work so that they can do the reasoning. They are basically switching between numerous cognitive tasks.

Below is a screen grab from a video of one technician conducting such procedures.


The problem

Field technicians show up to substations and other work sites without knowing how to navigate the site. They also need to find the proper tools and documentation for completing maintenance, tests and repairs to assets.

They continually circulate their eyes between consulting site plans, asking site staff where assets are located, monitoring asset performance and completing their task checklist. Their hands are likely juggling multiple items that they need to set down in particular locations and keep track of in order to do the job.



With augmented reality, field service workers don glasses at the job site. The glasses give them an internal map of the site they are at. Locations of tools and supplies are flagged or highlighted. Digital documents needed to complete a task appear in their field of vision. They can see any chart for an asset by shifting their attention to the asset. Field workers won’t need to remember anything or fumble with tools and documentation simultaneously. They can just focus on performing the work.

Here’s an example of how a substation needing maintenance might appear with augmented reality.


  1. The menu helps decrease the total space filled by UI elements. Only the current UI element is up. Others appear wilted until workers turn gaze at it or access it with their hand or voice.
  2. This check list helps the user complete their work.
  3. The radar helps them navigate the station. The orange square shows the location of the nitrogen they will need.
  4. A capacity indicator of the transformer cylinder is off in the distance. Relative size is used to indicate importance. It isn’t up close because it isn’t exigent and spatial relationship is the primary representation in use.
  5. The chart shows year-over-year performance for the transformer. The orange year shows the current performance which fluctuates outside the range it should be in.
  6. The problematic area of the transformer is highlighted to draw attention to the user

This is not science fiction. It is reality. All the technologies and features listed in this use case are completely feasible with many of the headsets coming on market in 2016-2017. One example I love is the Meta headset which has gesture tracking among other awesome features.


The Benefits Of Automating Your Job With Analytics


office collaboration tablet web

Automation and robotics are becoming increasingly capable of taking over at least aspects of many jobs. Some people are concerned that their entire job will replaced. However, there is good news if you’re among those concerned about losing their jobs to robot overlords – researchers at McKinsey found that only roughly five percent of current US jobs could be completely automated using existing technology.

If automation and robotics won’t replace your job, they will certainly change significant parts of your job, mostly for the better.

The McKinsey researchers examined the benefits to automating parts of a job by surveying 2,000 tasks commonly found in American jobs. They estimate that in 60 percent of current US jobs, 30 percent or more of the work activities could be automated using existing technology. That equals 1.5 days in your work week. Not only would this work be taken off your workload, it’s likely the work would get done more accurately or to a higher level of quality.

Unfortunately, this likely won’t be mean shorter work weeks anytime soon. But, it does mean that you can spend more time doing tasks that people do well, such as setting strategy, creating content, supervising and mentoring.

Take the example of an average marketing manager. Marketing managers are often highly skilled and highly compensated workers. The organization needs them doing the work for which they are best or uniquely suited. Roughly 10-15 percent of tasks in such a job could be automated, such as reviewing pricing decisions, estimating material costs for product production, and surveying industry literature for trends and competitive information. Automating these tasks could free up roughly half a day per week for a marketing manager.

An extra half-day per week equals more than 20 work days per year. That time could go to new product research, increased sales training, additional product promotions, and other activities that directly impact the bottom line, provide job satisfaction and benefit from a human touch.

Analytics provides the brains behind automating job tasks. First, analytics build lists of tasks through anomaly detection, failure analysis, root cause analysis and other techniques. Tasks might include maintenance work, repairs, corrective steps, scheduled activities, etc. Next, these tasks are then scored according to complexity, criticality, urgency, cost impact, risk of danger, likelihood of human error, and other characteristics.

Depending on these scores, tasks might be done automatically, done pending review of a human, initiated only after gaining permission from a human, or left to humans to do. The tasks needing human permission, review or labor are presented in a prioritized list to facilitate decision making and resource allocation.

By sorting work in this way, and then carrying out some of the work, analytics become a powerful assistant to workers striving to serve customers better while optimizing organizational performance.


(Image courtesy of nd3000 / 123RF Stock Photo)


Could ‘Pokémon Go’ Inspire Enterprise Productivity?



The world is going crazy for Pokémon Go. Nintendo’s stock value is making huge gains while masses of people are out hunting little ‘holographic’ critters. The technology isn’t new. Yelp has had a similar functionality out for 4 years called monocle. Yet this ‘Pokemon Go’ has made a ginormous splash. How? They used a technology to solve a pain point which is evidenced by their profitability so I don’t want validate that product vision here. Instead I’d like to answer another question which is “How can Augmented Reality help Enterprise on the same scale as the B2C market?”

The first part of the answer is that the motive to introduce AR to enterprise should not be about making a lot of money but rather helping a lot of brilliant people attain more.  Guy Kawasaki says, “The genesis of great companies is answering simple questions that change the world, not the desire to become rich.” The same applies within an organization eyeing new product offerings.

The second part of the answer is more involved so first I’ll discuss proof that Augmented Reality indeed helps brilliant people dive into a use case and then highlight some good User Experience design principles stemming from neuroscience that will make the use case come to life.

Augmented Reality (AR) is a technology that registers digital aspects onto the physical world around us. A stop light could be considered primitive Augmented Reality (AR). AR on Mobile Devices isn’t new but neither was the tablet before Apple made mountains of cash by designing it the right way. After Google Glass, there was a healthy dose of skepticism that everyday people would enjoy using Augmented Reality. Many would assume that if people don’t want it on their glasses with their hands free then why would they want to use a hand to hold the same information that they didn’t find useful?

Even as AR is in its technological infancy, Market guru Greg Babb explains how Augmented Reality reduced errors and time to complete task for wing assembly study presented by Paul Davies in conjunction with Iowa State and Boeing.


The chart above shows that wing assembly took significantly less time than traditional methods of assembly. There tends to be a major drive in Enterprise AR use cases to build applications for field crews and those assembling things. Meta CEO Meron Gribetz is on the record of saying he expects to throw away all the monitors in his office by next year and just work with AR headsets. This is compelling because Meta consists of developers, designers and scientists. That means knowledge workers would be using AR at their desk. This is either crazy or prophetic.

So what would an enterprise application look like? Well lets look at it through the lens of situational intelligence. Let’s consider the following use case:

Wind Turbine Use Case

A scientist at an energy company needs to run some prescriptive analytics for wind turbines. Government compliance and regulations have just been reformed and costly repairs and updates to the machinery has to be implemented on an accelerated pace. Heavy fines will be instituted for safety violations. Our scientist wants to use Matlab to run simulations on existing wind turbines to predict which turbines have a greater risk of breaking down, overheating or malfunctioning. He is think she can get more longevity out of gearboxes for yearly use within certain confidence levels.

  • He first speaks out loud saying,”Show me wind farms in central California.”
  • He sees a 3D map of several wind farms that he spins with his hands
  • The length of time he gazes at a certain region makes it slowly expand
  • He sees temperature on 4 wind farms with a high failure prediction
  • He increases severity of wind velocity changes in his data set and sees gearbox 4 fail
  • His gaze on the fourth gearbox chart line causes a line to appear animating towards the holographic turbine
  • He picks up the turbine, swipes the outer shell away and sees the motor spinning
  • He dictates notes about the turbine and recommends replacing the gearbox sooner than others


Our scientist is able to work faster in a more implicit manner without opening various files and programs with a mouse. He has more visual real estate and can now work with depth rather than just up and down dimensions. He is more productive, which makes his company more money.

Neural Interface Design

Pictorial Cues are an important aspect of good user experience that make a scene more realistic in AR. Some of these can be:

  • Occlusion – One object blocking part of another
  • Relative Size – Equal sized objects taking less area within our field of vision when their distances from us vary
  • Shadows – Objects that casts shadows seem more real
  • Accretion – Aspects of an object appear as a user moves in the physical world

Another aspect to design for in AR is binocular disparity as it assists with depth perception. Studies show that the neurons in our visual cortex fire optimally when there is a amount of specific disparity with a stimulus. So that is why you see one wind turbine moved to the right of another in the wireframe.

So maybe you won’t get to go hunting Pokémon at work if your company is smart enough to outfit you with one but you will definitely feel like a Jedi when manipulating the world around you with your senses. You will love sharing this world with your co-workers. Augmented Reality unleashes you imagination. The next couple years show some exciting times ahead for Reality Computing!

Special thanks to JulianHzg for making a great Wind Turbine in Blender at  that I recolored and used in my wireframe.



Pipeline Analytics Lower Natural Gas Risk


pipeline welding web

Pipeline accidents in Allentown, Pennsylvania in 2011 and Sissonville, West Virginia in 2012 destroyed homes, caused death and injury, and reminded us how critical careful gas transmission and distribution really is. We are accustomed to natural gas reliably powering our homes and businesses, but even those responsible for getting it there can take that system for granted.

Safety and aging infrastructure are top concerns of natural gas executives surveyed by Black & Veatch in 2015, and with good cause. Replace just five percent of a room’s air with natural gas and the atmosphere becomes explosive. According to a U.S. Department of Transportation report, nearly one-third of natural gas distribution pipelines in the U.S. were built before 1970.  More than 50,000 miles of these older pipes were welded together with outdated techniques that are prone to failure.

Despite these known conditions and obvious hazards, one-third of respondents to the Black & Veatch survey did not have a resilience plan in place for their natural gas operations four years after the Pennsylvania and West Virginia accidents. Fifty-four percent of respondents agreed with the statement that “a formal risk-based planning approach has not yet been undertaken to my knowledge.”

Why should so many utilities executives be without a risk-based plan, when the consequences of risk are so high?

Much of our natural gas infrastructure is hidden underground, which means it is often out of sight and thus out of mind, contributing to our becoming inured to problems. Gas utilities have developed clever methods such as pigging, hydro testing, and cathodic inspection for measuring and maintaining the health of pipes. Those clever methods generate multiple, disparate sources of data related about assets. Today, many utilities are flooded with data but no closer to fresh and useful insights based on that data.

Situational intelligence offers a powerful approach to analyzing those data sources and quantifying the risk present in natural gas assets. By correlating, analyzing and visualizing data related to an asset’s age, condition, location, network relationships, and operating history, situational intelligence provides a method for making decisions based on the likelihood of asset failure and the consequences should failure occur.

With this specific understanding of risk, natural gas managers and executives can prioritize maintenance, repair, refurbishment and replacement work to focus first on the most critical assets. This approach drives down risk faster than following a time-based or even condition-based approach to asset planning and operations.

We can’t fully eliminate risk, but we now have the analytics approaches to understand, quantify and lower risk to help prevent future pipeline accidents.

(Image: smereka / 123RF Stock Photo)


Three Ways Analytics Improves Emergency Response


fire office building--web

In previous posts about safety, I’ve talked about how predictive analytics help prevent accidents before they happen. But, analytics are also useful for improving the speed and effectiveness of response when accident do happen.

Here are three ways to apply analytics to emergency response:

Analytics help you identify the real problem

In today’s IoT and Big Data world, you may have multiple systems creating alarms in response to a single accident or event. How do you know what the real problem is? Analytics that bring together alarms from across your enterprise help identify and respond to the real problem.
For instance, your building management system gives you an alarm that sprinklers have turned on in one building of your corporate campus. Does this mean that there is actually a fire, or that someone created a false alarm, or that the sprinkler system has somehow malfunctioned? Without analytics that correlate the sprinkler alarm with other data such as smoke alarms and security video, it’s hard to know.

Analytics help you triage effectively

Alarms tell you that something happened, but by themselves don’t tell you the consequences of what happened. Analytics that include a criticality score for locations, equipment and inventory quickly give you the true magnitude of an accident or event. Knowing the magnitude of consequences also allows you to prioritize your response to multiple simultaneous events.
Continuing with our building fire example, let’s say that a correlated alarm system tells you that there is, indeed, a fire. You immediately want to know the potential consequences of the fire. How many people work in the area? What sort of equipment or inventory is nearby? Are there any guests in the building? Without analytics that rate the consequences of the fire, it’s hard to fully assess the situation.

Analytics help you respond efficiently

Once an accident or event has been properly identified and assessed, there’s no time to waste with inefficient or ineffective response. Analytics that correlate the type of incident with the specific qualifications of first responders and their respective current location means that people arrive on the scene ready to act, instead of ready to assess. The same approach applies to vehicles and equipment that first responders may need to address the situation.
If our building fire is a chemical fire in an inventory warehouse, that situation requires a different type of response compared to an electrical fire in an office building. Without analytics that correlate people, equipment, locations and events, you risk having the wrong people respond with the wrong equipment to handle the situation.

(Image: stanislaw / 123RF Stock Photo)


Are You Ready For Autonomous Freight Trucks?


truck convoy web

Recently a group of freight trucks drove across Europe autonomously. Daimler is now selling new semis with autonomous driving features. A Silicon Valley start-up wants to sell equipment to retrofit existing freight trucks for autonomous driving.

Are you ready to share the highways with autonomous trucks?

Autonomous trucks are not driver-less trucks. Trucks still need drivers to handle city driving, parking, refueling, delivery paperwork and many other aspects of freight handling. It’s during those long, boring stretches of highway driving where autonomous systems relieve drivers of some of the dullest and most dangerous work.

The work is dull enough that there’s currently a driver shortage. The American Trucking Associations report that the United States needs about 50,000 more truck drivers. Autonomous systems might help make the job more appealing, and maybe even make today’s drivers more productive. Both outcomes would help with the driver shortage.

Autonomous trucks are designed to have fewer accidents, more predictable performance, and better fuel consumption. One way these trucks conserve fuel is by lining up in platoons to reduce wind resistance.

The trucks use a combination of video cameras and radar to sense the situation around them. That video and radar data can be enriched with other information about weather, terrain, destination and cargo, then analyzed to optimize routes, schedules, fuel consumption and more. The analysis may suggest, for example, that the truck is ahead of schedule and slow the vehicle to arrive on time and conserve fuel in the process.

If your company owns a fleet of such smart trucks, then you might benefit from analysis spanning groups of trucks or your entire fleet. For instance, could you change schedules and routes so that more of your trucks travel in platoons to save fuel? Will the suggested changes to routes and schedules still meet your shipping agreements or delivery commitments?

For years as air travelers, we’ve ridden in planes with autonomous piloting and enjoyed it as the safest form of transportation. But when we’re operating our cars alongside autonomous trucks, does our attitude change? Personally, I’m not sure how I’ll feel passing, or being passed by, a seeming wall of platooned trucks rolling down the road.

What are your thoughts?

( Image: whitestar1955 / 123RF Stock Photo )


Analytics, Transportation Safety And Extreme Weather


transportation safety landslide 01

Extreme weather conditions such as floods, landslides, sinking water tables, droughts, ice storms and snow drifts takes a toll on the landscape. Those weather-driven changes to the landscape directly impact rail beds, roadways and bridges, rendering them unsafe to use until conditions change and repairs are made.

For instance, heavy rains trigger landslides that block roadways and railways, stopping traffic until the debris is removed and the damage repaired.

Analytics can highlight where and when disasters might strike. Predicting where damage may occur and taking precautions prevents damage to people and property.

Factors that signal a landslide or flood hazard include

  • Transportation corridors in proximity to sloped terrain or bodies of water
  • Vegetation, or lack thereof, on slopes and banks
  • Moisture levels in the soil
  • Past, current and projected weather conditions
  • Condition of the road or railway itself
  • Traffic metrics for the corridor including overall volume of traffic, patterns of traffic flow, and criticality of the corridor in connecting valuable locations

Imagine a situational intelligence approach to transportation safety. An analytics application could correlate, analyze and visualize the factors listed above. That analysis and visualization would enable government transportation officials and railroad operations and planning professionals to see the probability of landslide on a slope adjacent to a right of way. Knowing the probability of a landslide and the magnitude of its impact informs decisions about operations, crisis response and mitigation.

A similar situational intelligence approach applies to floods damaging transportation corridors that are adjacent to bodies of water. Some of the measurements and algorithms would be different, but the much of the output and outcomes from such a system would resemble that from landslides.

When public transportation departments and private transportation companies know that a corridor is at risk, they can take action to avert disasters. Traffic and shipments can be redirected to minimize delays and remain safe. Customers can be notified if their shipments will be delayed and new arrival times. Repair crews can be positioned for faster response to the most critical areas.

Analytics for transportation and weather is especially important in the United States given the current condition of transportation infrastructure. The American Society of Civil Engineers give these systems low marks:

Infrastructure that is in barely passable condition is less resilient to the impact of extreme weather. If we can’t change the weather and don’t have the money to improve the infrastructure, at least we can be smart about how we plan and respond.

(Image courtesy of fotokostic / 123RF Stock Photo)