AndrewNairn
IS-IT--Management
I am trying to specify a method for historically analysing data collected from the machines that we make which we can supply as an extra to customers, and also use it as a means of troubleshooting our machines when there is a service call.
The machines deal with metal processing, and currently have about 300 measurement points collected by an omron PLC and stored on some flash memory. The machine is controlled by a touch screen attached to the PLC.
I am looking at some of Wonderware's offerings (IndustrialSQL and ActiveFactory).
The system needs to:
- Store data for a reasonable period of time and allow fast access to it.
- Allow trend analysis and low-level alarms (email or similar) if something is showing pending problems.
- Be easy for the customer to use (i.e. they don't need a full-time programmer to maintain it).
- Not require a lot of post-manipulation of data to get meaningful results.
- Allow us to access the data remotely to troubleshoot. The machines are often in remote parts of the world where full internet access is problematic, and customers can be secretive about their production data.
Questions:
- Is a wonderware solution overkill for what we need (i.e. will customers pay for it...) when there may be a much easier way of achieving what we need?
- Anyone use anything similar in terms of trend analysis and what are their experiences?
Any advice warmly received!
Thanks, Andrew
The machines deal with metal processing, and currently have about 300 measurement points collected by an omron PLC and stored on some flash memory. The machine is controlled by a touch screen attached to the PLC.
I am looking at some of Wonderware's offerings (IndustrialSQL and ActiveFactory).
The system needs to:
- Store data for a reasonable period of time and allow fast access to it.
- Allow trend analysis and low-level alarms (email or similar) if something is showing pending problems.
- Be easy for the customer to use (i.e. they don't need a full-time programmer to maintain it).
- Not require a lot of post-manipulation of data to get meaningful results.
- Allow us to access the data remotely to troubleshoot. The machines are often in remote parts of the world where full internet access is problematic, and customers can be secretive about their production data.
Questions:
- Is a wonderware solution overkill for what we need (i.e. will customers pay for it...) when there may be a much easier way of achieving what we need?
- Anyone use anything similar in terms of trend analysis and what are their experiences?
Any advice warmly received!
Thanks, Andrew