"IDS's goals are to help oil companies transition to the digital oilfield, provide ANOVA-as-a-Service to enable true knowledge sharing, deliver best practices, and grow the customer base," says Chief Technology Officer Reuben Wee. "The major challenge is data quality. It's a concern because every client has their own best practice codes, data format, and data structure, and we needed to make sure we could correlate various best practices and align them for benchmarking. "We also have to look at the secondary data channels proposed by our clients. In data analysis, if there is any doubt at all as to the accuracy or integrity of the data, it is just not usable. If we weren't able to resolve these data quality issues, the benchmark reduces in value.

"Our technology initiative was to find a tool that would help us tell the data story, the facts of their data, but told in their functional design or industrial language. These narratives allow clients to quickly learn and inject those lessons back into their workflow.

"Traditionally, all documents and data from a drill rig go to one person who keys it in. One of our applications, Lean Automated Reporting (LAR), receives data from all the sensors during drilling, and from a state detection system we devised, and converts it into actual human readable reports. It uses machine learning, for which you need a curated analysis of variance (ANOVA) data source to translate noise to signal. LAR reduces human error by reducing human contact with the data. It's essentially a more truthful reflection of sensor data, a more transparent way of showing how you derived the data.

"ANOVA may need structured databases, NoSQL databases, even flat files. Every client is different. About three years ago, we decided to step up our game and look for a solution to see us through the next growth stage. We needed a tool to be able to bring these different data sets together and align them by either time or depth. It had to be easy to learn, configure, and integrate. It had to be flexible and scale properly. We found that Jaspersoft® was the best fit for knowledge sharing among tools.

New Capabilities, New Clients

"Jaspersoft has some key features, the ETL tool that is already bundled in, a server that provides the scalability we require, and Jaspersoft® Visualize.js™ that allows us to tightly integrate dashboards and the reports into the application.

"Because of the current industry slump, we are looking beyond traditional data collection and processing customers. We now have clients for whom, instead of collecting data, we're looking at it. We attribute that to our technology behind ANOVA. Our sales team can propose solutions to a broader set of problems. We expect to see a 50 to 100% growth in new client base just from this service alone.

Agility That Wins and Keeps Business

"Business requirements change quickly, and the expectation is that we will provide answers, dashboards, and reports quickly. One of the main differentiators is Jaspersoft: the ad hoc analysis tool and Jaspersoft® Studio to quickly develop dashboards as a proof of concept. If that works, we go into specialized graphs and charts that are more attuned to the industry language.

Fast, Efficient Story Telling

"Jaspersoft helps us to be better storytellers. It provides a platform, a canvas to bring the data together and quickly and efficiently give the clients insight into data through visualizations.

Reduced Workload, Greater Innovation

"Because of the simplicity of the tools, we've reduced our workload by nearly half, freeing up time for the developers to look at innovative ways to do visualization. We've always been a lean, close-knit team, but Jaspersoft has allowed us to do more innovation using a third-party library such as INT or D3.js. We do so confidently knowing that the final result can be rendered in our application as well as integrated into reports and dashboards with Jaspersoft.

"Up to now, ANOVA is used in the back end, where it processes data and turns it into tangible reports and visuals. We are going to take this knowledge and feed it back into operations to achieve continuous situational awareness with Lean Automated Reporting. That will reduce the need for humans to be involved in the process of regulatory and operational reporting. It requires machine learning and processing of live streaming data, and possibly will lead to managing events, as well."