“Big Data” is all the buzz in Information Technology circles and business editorials. Generally the discussion is around how we will ever be able to make meaningful use of all the vast, undefined data that is being created.
It’s all hype. Software vendors want to add urgency to the adoption of computer systems that manage business information. It’s a great message if you are one of many high end IT companies trying to sell customers on services. Unfortunately for customers, most IT company’s messages are non-specific generalities that could be applied to anything.
There is a huge disconnect between the IT world and specific applications of computer science. In order for the programmer to produce “good” code, good functionality, an application that has value for the customer, the team writing the code must include a domain expert in the field the code is being applied to.
Yes, we have the ability to extract mountains of information about anything. For example, when Device Net was first released it contained a list of 168 parameters for starting an electric motor. This was incredibly frustrating for early users of the technology because the first units had no factory default settings, each parameter had to be programmed or the starter would not operate. Major pain.
A group of programmers can come up with hundreds of data fields because they are thinking in terms of information for it’s own sake. In some respects, software products generates revenue by producing more data. This is clearly the opposite of what is required in most applications, reducing the information to only what is relevant and necessary to a specifc application or field of use.
For each specific application of the starter, only a handful of parameters will actually be meaningful. When was the motor started? (day, date and time tag) Was the current profile within acceptable values for the specific motor horsepower? What are the total number of hours that the motor was running? Did a current imbalance take place? (day, date and time tag) Was there an undervoltage or overvoltage event? (day, date and time tag)
The most valuable applications of so called “Big Data” in the manufacturing world are centered around Operational Effectiveness and Predictive Maintenance. It’s about cost control from a business overall business perspective. As manufacturers merge IT and Manufacturing it is inevitable that every modern piece of control equipment will be an Internet Appliance.
The future is upon us.