In manufacturing plants, business applications are playing a major role in making decisions effectively in real-time. Management would like to be informed on regular intervals to drive better operational efficiency & revenues. It’s always said “What you can't measure, Can never be controlled” likewise in production plants, if you cannot measure the production attributes in real-time then it would be extremely difficult to optimize and improve the efficiency. Lack of real-time production leads to poor analysis and decisions may not be effective. In competitive markets and growing economy, if one doesn’t make “Optimal utilization of production plant” then this will be nothing less than a crime.
To run manufacturing plants in optimal mode, we need to collect the data from production floors and to make better decisions, we need data that’s more Authentic, Accurate and on the real-time basis. Any guess? Where do we find such data? Yes! You got it right; this data is hidden in the controllers of each machine in the shop floor. All we need to do is, connect and collect data from these machines in a real time mode.
In any plant, there are two types of machines new generation and legacy systems.
New Generation: Getting data from new generation machines is easy and most of these machines already have built-in “Open connectivity” protocols. These protocols support a quick and easy way of data exchange mechanism. Please do check the specification of new machines while ordering and if they aren’t supporting them, then it’s recommended that one need to ask for the open connectivity. It’s based on the machine supplier, some do provide and some will still continue to have their own proprietary system protected from any sort of connectivity...
Legacy systems: Getting data from the legacy system is a bit of challenge sometimes. As they many not support open connectivity or they may have proprietary protocols and getting details from these vendors is most often a time-consuming activity but never an impossible task. We need to study the machine details, understand its communication architecture and explore the possibility of add-on connectivity. Here are the quick 5 steps for understanding:
1. Existing Communication Channel: Do we have communication channel or port accessible in the existing control system and if not, can we add a communication port to it. The addition of communication module will depend on various PLC/SCADA brands and it also depends on various data sources.
2. Connect: let’s say we have communication port available, then we need to understand its communication protocol ( eg. Modbus) based on this we need to aggregate similar machines and make a network of machines. One also needs to understand the distance between the machines, as this would be one of the key aspects of building physical network connectivity.
3. Access: After establishing the connectivity, with the help of programming software we need to check if we are able to upload the program from the existing controllers. This is where some of the machine builders will lock or password protect their controllers. We need to talk to the OEM (Original Equipment Manufacturer)to get the password to unlock the system and then use it for data exchange.
4. Identify data exchange points: Once connected to the controller, we’ll be able to send / receive commands successfully, now let’s identify the key data points from controllers. Eg. Start time, stop time, quality rejection, tool changing time, OEE, etc... it’s important to have a discussion with the production team and identify the key process parameters that we would need to extract from the given machine.
5. Aggregate: Once data points are ready for each machine, map the data points to the centralized location and aggregate all the data. You can build the aggregation system using an existing system (SCADA or OPC based system, historian application) as well. OPC ( OLE for Process control) is the ideal choice if one needs to install a new system or as we have seen historian can also be used based on needs. Once we have data aggregated in a historian, you are empowered with data and it’s a raw data only. To convert data into information, we need to do the thorough analysis of each stamp of data. This can be done by generating and monitoring key production reports on regular basis.
Now that complete control of data or refined data is available, this data can be feed directly to Enterprise applications like ERP (SAP) systems. An ERP system helps us to understand and derive more value in determining the actual production cost & utilization of machines in real-time. As superman says “Great power comes with great responsibility”. Real-time data gives us the power and basis for making decisions in real-time. Yet, understanding the “process impediment” and resolving them will be our greatest responsibility.
