Maintaining Data Continuity When Networks Fail

Published on: 

Equipment and Processing Report

Equipment and Processing Report, Equipment and Processing Report-04-18-2012, Volume 0, Issue 0

My network failed and I had to scrap my batch because my historian did not collect the required data. How do I upgrade the reliability of my network to maintain data continuity if my network fails again?

Q. My network failed and I had to scrap my batch because my historian did not collect the required data. How do I upgrade the reliability of my network to maintain data continuity if my network fails again?

A. Data collection for regulatory compliance is a crucial function for pharmaceutical manufacturers that makes network reliability and data continuity of upmost importance. When reviewing the reliability of a network, first consider incorporating redundancy into your connections and switches. Added redundancy increases network resilience in the event that a switch or connection fails. Second, work closely with your company’s IT group to proactively monitor network health within the process control system. Managed switches have the intelligence to detect issues, such as when cables are broken, switches are powered down, or network loops are accidently broken. Another way to identify potential problems is to use situational alarming. Process engineers are able to acknowledge alarm events that might jeopardize the network and use the information they receive to identify and address issues almost immediately. Third, review network configuration against your IT and control system vendor recommendations. Proper network configuration is crucial to preserving your network’s security and resiliency in the event that a failure occurs.

Building a redundant, fault-tolerant network as described above is a dependable but expensive solution. An economical and reliable alternative is to collect the data at the equipment level, in the process control-system rack, and push it up the network to a site-level historian when the network is available.

Equipment-level historians are designed to stand alone, with limited software footprint and no server connections, which makes the unit inexpensive to deploy and helps reduce the risk of data loss due to network or other system interruption. The unit’s onboard memory maintains a continuous data buffer that allows it to offload and forward that data up the network when the connection is available. This is especially important with original equipment manufacturer skids that may have a latent connection to the network. Another benefit of equipment-level data collection is the ability to collect data at higher speeds, allowing for a more granular view into your process. For example, traditional historian events are recorded every 2–3 s, but collecting data in the rack allows you to record process parameters as frequently as every 10 ms. This high-speed sampling rate can enhance investigations and batch analysis, which can ultimately improve product quality and speed time to market, thus increasing production volumes.

Advertisement

—Jason Wright, PlantPAx product manager, Rockwell Automation.

If you have a problem with your equipment or process, an industry expert may have the solution. Please send your question to Jennifer Markarian, editor of Equipment and Processing Report, and we may be able to provide an answer in a future issue. All questions will remain anonymous.