We are rapidly moving into an age when every machine and piece of equipment, from the very smallest to the very largest, will be capable of collecting massive quantities of data, often from multiple sensors. Those data can be delivered to someone who needs to make use of them for diagnosis, performance improvement or any of a wide range of other purposes that are enabled by the availability of that information.

The potential this opens up for making better use of our resources is immense. But there is a problem: As is now so often acknowledged in the technical press, we are drowning in data. The amount of data is just too large to make sense of.

In fact, the problem is not the data themselves. We ought to see the data as a solution rather than a problem. The real problem is the shortage of the expertise required to make sense of the data and to communicate the meaning of these data to those who need to act. This problem is widespread: Numerous studies across a variety of industries have bemoaned the acute shortfall in human resources that prevents us carrying out analysis of the data and delivering the information that they contain.

Natural language generation (NLG) provides a solution to this problem. ARRIA’s NLG Engine is a sophisticated software solution that combines cutting-edge technology advances from two areas—data analytics and computational linguistics. The software essentially embodies the expertise of senior engineers in analyzing the data to extract useful information and in communicating that information in natural language reports that read just as if they had been written by those engineers. But because it’s a software technology, it’s easily replicable and scalable so that it overcomes the human knowledge resource bottleneck. And it’s also extremely fast. Analysis and communication tasks that can take hours for a human engineer to carry out can be achieved by the software in minutes.

ARRIA’s NLG Engine has been used to analyze and communicate information in the health sector, in meteorology and in finance, but the technology also is in use in the oil and gas industry. Shell uses the NLG Engine for rotating equipment surveillance across six deepwater oil and gas platforms in the Gulf of Mexico (GoM).

Shell’s deepwater collaborative operational and engineering surveillance decision support center, the Bridge, and its exception-based surveillance (EBS) technology platform provide an environment where Shell combines and integrates tens of thousands of data streams from multiple sources for all of its deepwater assets in the GoM and South America through one common IT framework. These data streams are bundled through predefined calculations in an analytics engine into critical conditions of interest that are monitored 24/7, 365 days a year. Shell then looks to prevent, treat and remediate any emerging issue flagged through its automated analytics EBS technology platform.

Until recently, this process required expert (and scarce) engineers to analyze the situation to come up with a recommendation for action. This can be a time-consuming exercise, sometimes taking three or more hours to dig through the various data streams and maintenance histories of the affected piece of equipment and of related pieces of equipment to work out what might be going on.

Systems like the NLG Engine can do this same work in 60 to 90 seconds, producing a natural language report that is indistinguishable from those written by human engineers. This result is achieved by combining sophisticated analytics capabilities that emulate the engineer’s reasoning with articulate communication capabilities that emulate the engineer’s ability to explain what is going on in language that is relevant to the operator.