Welcome to the ‘Introduction to Data Collection’ webinar.
Leading the session today is Robert Kaye, a Product Marketing Manager here at Atlas Copco. In this webinar, Robert is giving an introduction to data collection, describing who needs it, why you need it, and how to get it. We begin the webinar with this fact: big data and data analytics are the current trends in manufacturing industries today.
Building off of this, Robert explains data silos, which is the first challenge we must solve when it comes to data collection. Data silos are isolated groups of data that haven’t been processed or analyzed. They usually take the form of raw data. This data is inaccessible and unreadable. A key feature of Industry 4.0 is making that unreadable data accessible to the people in your plant that need it.
On that subject, Robert begins talking about who it is that actually needs the collected data. Maintenance, Manufacturing Engineers, Management, and Customers are the people for whom the data is most valuable. How do we collect this data? There are multiple ways. Physical access refers to someone walking to each controller and collecting the data. One to one is when you connect to one controller at a time via the plant network. Automated data collection refers to controllers sending data to a shared site with little to no user interaction.
Following this, Robert explains how you can start the data collection process. You must have a network, using either Ethernet, WiFi, or a Fieldbus. There needs to be an embedded communication protocol such as Open Protocol or TNOP. You need a database structure and logic to store/organize data, and that can come in the form of a SQL Server, an Oracle, etc. And finally, you need a reporting mechanism, such as a Windows client or a website.
In the final part of the webinar, Robert moves on to some key elements of data reporting. The reported data needs to be usable, and oftentimes, it should be customizable for specified departments or user needs. Reporting should not be static or reactive, it should be proactive. It should drive quality and increase efficiencies while easing any containment issues that occur. Your data collection service should have a proactive alerting feature that sends out conditional alerts directed to people who can act on them quickly.
Robert brings up one final question: should you build a custom data collection solution or buy one off the shelf? There are pros and cons to both. If you’re looking to build a software solution for data collection from the ground up, you should know that the major pro to doing this is that it’s a completely tailor-made solution. Cons include longer lead times, never-ending support, and resources that can cover building software. For buying off the shelf, the pros are that it’s a proven solution with long-term support and known costs. Cons are that there is a long lead time for change requests.