Industrial wireless is based on mesh networks, which are well-suited to batch applications that are subject to being reconfigured, since not only do wireless networks provide bidirectional digital communications, but they also have the flexibility to continue to work as the environment around them is changed. Another benefit of wireless is that the network does not require a penetration through a barrier to communicate, provided the barrier is not metal, so it would work well in a clean room or glass bioreactor.
An active or passive RFID sensor can transmit from within the reaction vessel to an 802.15.4 or 802.11 gateway outside the process, but within meters of the signal source, where it can be received and propagated onwards.
For more data-intensive operations, battery-operated nodes can gather or aggregate data, send it and then sit in sleep mode for long periods. The ability to sleep and to be located practically anywhere works especially well for sensors that frequently need to be relocated, as is the case in the batch facilities widely used in the pharmaceutical industry. The pharmaceutical industry also has to meet good automated manufacturing practice (GAMP) requirements, where more and easier access to data makes meeting the documentation requirements that much easier.
A key component of GAMP is process analytical technology (PAT), which requires data to be integrated from chemical, physical, microbiological, statistical and risk analysis. In addition, PAT is used broadly in batch manufacturing implementations facing many challenges, and so far, few facilities have documented online applications.
A recent article by Terry Blevins, et. al., “Batch Operations Benefit from Process Analytical Technology”, described the requirements for a successful application of batch analytics for online operation. A key component for PAT to be successful is that the system has to be flexible, yet robust enough to accommodate the following operating criteria:
- Process holdups;
- Access to lab data: integration of LIMS with the process;
- Variations in feedstock;
- Varying operating conditions;
- Concurrent batches — multiple batches of the same product may be executing at various stages of completion;
- Assembly and organization of data — requires that it can be accessed, correctly sequenced and organized in a meaningful way;
- Data alignment from different batches — batch durations are not equal, so forming data with an equal number of data samples for every batch is important for consistent data analysis across different batches.
Once the data identified above has been gathered by the wireless field sensors and associated networks, it must be transmitted from the batch equipment to the central control system or databases, so it can be properly captured, historized and manipulated to provide the information listed above on demand.
One way to extend distances is by creating either IEEE 802.15 or IEEE 802.11 mesh networks, in which nodes send data to a neighboring module that will relay it to its final destination. When properly configured, mesh networks are fault-tolerant, since new routes can be created when a node fails. However, the penalty, if there are multiple nodes or hops, is lag in getting the message through. The most realistic approach to sending data over long distances is to convert the gathered data via a gateway to IEEE802.11s, and forward the message(s) to the central data repository using the greater distances and bandwidth of Wi-Fi.
The pharmaceutical industry is secretive about how it is using new technologies, but I am confident that if wireless is not being used now, it will not be long before one or more of these companies realize that wireless can help them move closer to their reliability and profitability goals.