IIoT PLATFORM FOCUSES ON SCALABLE ANALYTICS
As the world increases its need for connectivity, oil and gas producers have the need for people who understand data. According to Gavin Rennick, president of software integrated solutions at Schlumberger, speaking at Automation Perspectives, the best results were obtained when there was a successful connection established between people who understood the technology along with those who understood the domain. Automation Perspectives is Rockwell Automation’s media event that leads up to the Automation Fair. He added that it sounds simple but the amount of work that can be done in such a short period of time, is simply amazing.
At another event focusing on modernizing water systems, a professor at the University of Michigan called on a bunch of students perched at the back of the room. Even though their focus lay on environmental engineering and studying water quality, the students were pursuing a dual degree in computer science – which will help them to use the tools available and make sense of the data collected.
While there is a need for data scientists in all kinds of industries, there is a continuing trend in the number of rollouts from automation suppliers that are trying to make data useful and more accessible to the domain experts. FactoryTalk Analytics works towards helping industrial analytics overcome the complexity of data and makes it easier to combine structured and unstructured data from any virtual source and derive meaning from the data with natural language searches that make sense to domain experts.
With increase in connectivity and good computing power, there is no longer a shortage of data. But to make that data useful for running and maintenance of ongoing operations isn’t easy. Project Scio reduces the hurdles and gains access to actionable information to fuse the data and deliver intelligent analytics into intuitive storyboards. Users can perform self-serve drilldowns and reduce time to value.
What are scalable analytics? A dimension of scalability is the type of analytic. Descriptive analytics tell you that the motor has stopped but diagnostic analytics will also provide you with the reason. By making use of resources available from IIoT, there is a need for predictive and prescriptive analytics – what is going to fail and what can be done to avoid it.
Data scientists that are trained to make better sense of data spend about 60% of their time cleaning and filtering useful segments of it. The ease of use for such scientists is going to be a very important concept required by the customers.
Read More Articles :
- French Muslims to create basis to finance mosques
- Tech For Supply Chain Finance
- 5 COSE CHE (FORSE) NON SAI SUL TUO MAC
- One trader has a strategy to win big from the Fed’s rate ‘ping-pong match’
- GOP objectives Dodd-Frank finance rules
Andrew Ellis, the global commercial engineering manager for information solutions at Rockwell used the example of a fictitious customer to show the ability of options for visualization and different charts that could be downloaded for presentations. He showed how efficiently all the data was drilled down to get information about one of their customer’s plants. He took the position of the plant’s operations manager and wanted to reduce energy. By using the service, he saw the kind of product that was using more energy through an intuitive storyboard.
He drilled further down to the level of a line operator and took a closer look as to how each one of them performed in relation to the energy consumption. Different operators run the line differently and the most efficient were the ones he was interested in and how they would do their jobs. He could tabulate the data by time rather than the product to figure out how the energy was consumed from changing shifts by operators or from the machine itself. A domain expert would be puzzled to come up with a solution and reduce energy consumption and not know how to summarize the data. By typing a search query, he was able to create data sets at the click of a button.
The key attributes included here are device auto-discovery, central location of all the data that can be continually refreshed, flexible machine learning, closed-loop analytics, open platforms and an application marketplace that can be introduced for in-house and third-party application development.