When can you skip thermal validation in a regulated environment? Paul Daniel, Vaisala Senior GxP Regulatory Compliance Expert Published: Mar 5, 2014 Industrial Measurements Life Science Wait just a darn minute... You can skip validation/mapping in a GxP environment??? Well... no... yes... maybe. It depends!!!
I'm curious if you have ever used continual monitoring to justify not validating a process. e.g. A refrigerator that is being monitored 100% of the time does not require validation in itself. HI K, Thank you for writing! There is a regulatory expectation to demonstrate that you understand your storage environment and that you have a high degree of confidence in the temperature uniformity of that environment. This is usually solved by mapping, then monitoring. However, it is theoretically possible to monitor, then map! Please note that you aren’t skipping the mapping validation – you are doing the mapping validation using the monitoring system as the mapping tool. Let me explain this process, which I like to call “continuous mapping." There are several factors at play here: the cost of the instruments their maintenance cost (recalibration) and the labor cost to deploy and retrieve them As the cost of instruments goes down, and the recalibration cost decreases (or the calibration interval gets longer), and as labor costs go up – this can be a viable option in some circumstances.Having said that, if/when I would encourage doing this, it probably wouldn't be in a refrigerator. The density of sensors needed is just too high, and the labor of deployment and retrieval so low, that it probably won’t ever make financial sense to do. (Proviso: see my comments at the end of this blog where I mention disposable devices.) ow with a warehouse, this could make sense! Consider this – in a 20 cubic meter cold room, I will use 16 sensors. In a 40,000 cubic meter warehouse, I will use 50 sensors for a fairly conservative mapping. My relationship of volume to sensors is logarithmic. Let’s consider that of those 50 sensors, I probably would have 10 in a monitoring system, so we are looking at an “opportunity cost” of 40 sensors. sensors are $1000 each and cost $200 to calibrate, I have a $40k initial investment, and a $8k annual maintenance costs. If I re-map every three years, and I do summer and winter mapping, over a 10-year period, I will be mapping 8 times. My system cost over the same time period will be $120k. That works out to $15k per mapping session. Does that sound like a bargain? To get a mapping that is guaranteed to capture the hottest part of the summer or the coldest part of the winter? What if the sensor price and calibration costs dropped, say to $500 per sensor and $100 to recalibrate. Now our 10-year cost is $60k. Each mapping session is now $7.5k. Does it sound like a bargain now? So now the math is simple. At what sensor cost (and recalibration price) does it save money to leave the system in place for continuous mapping? A difficulty also comes in selling the idea, as the limitation may likely be the upfront cost for all the sensors. Monitoring systems are usually considered “Capital Expenses,” whereas validation and recalibration costs are usually considered “Operating Expenses.” So you really need a situation where management is focused on total cost of ownership, not just Capital Expense represented by the acquisition cost. As long as sensors and recalibration decrease in price, “Continual Mapping” will eventually make sense, especially in areas with high labor prices, or frequent remapping. But, let’s not forget new technology! Disposable loggers used for shipping purposes are currently available for $40 USD or less. They often have a +/-0.5C accuracy and a battery life that is good enough for at least 1 month of mapping. Now my 50-point mapping of the warehouse has a hardware cost of only $2000. I think these disposable devices will become the standard for small chamber mapping – refrigerators, freezers (except ultra-lows), and incubators. I also think continuous mapping will eventually make sense in warehouses. We will probably be sticking to some level of traditional mapping when we discuss areas where humidity is concerned, or when we deal with extreme temperature or pressure (lyophilizers, autoclaves, ultra-low freezers). I hope this answer was useful, and that you will send us any future questions you might have! I am speaking at both Interphex NYC and the Measurement Science Conference in Long Beach in March 2014. I’ll be talking about temperature mapping. If you have a chance, please stop by and say hi! Best regards, Paul
I'm curious if you have ever used continual monitoring to justify not validating a process. e.g. A refrigerator that is being monitored 100% of the time does not require validation in itself. HI K, Thank you for writing! There is a regulatory expectation to demonstrate that you understand your storage environment and that you have a high degree of confidence in the temperature uniformity of that environment. This is usually solved by mapping, then monitoring. However, it is theoretically possible to monitor, then map! Please note that you aren’t skipping the mapping validation – you are doing the mapping validation using the monitoring system as the mapping tool. Let me explain this process, which I like to call “continuous mapping." There are several factors at play here: the cost of the instruments their maintenance cost (recalibration) and the labor cost to deploy and retrieve them As the cost of instruments goes down, and the recalibration cost decreases (or the calibration interval gets longer), and as labor costs go up – this can be a viable option in some circumstances.Having said that, if/when I would encourage doing this, it probably wouldn't be in a refrigerator. The density of sensors needed is just too high, and the labor of deployment and retrieval so low, that it probably won’t ever make financial sense to do. (Proviso: see my comments at the end of this blog where I mention disposable devices.) ow with a warehouse, this could make sense! Consider this – in a 20 cubic meter cold room, I will use 16 sensors. In a 40,000 cubic meter warehouse, I will use 50 sensors for a fairly conservative mapping. My relationship of volume to sensors is logarithmic. Let’s consider that of those 50 sensors, I probably would have 10 in a monitoring system, so we are looking at an “opportunity cost” of 40 sensors. sensors are $1000 each and cost $200 to calibrate, I have a $40k initial investment, and a $8k annual maintenance costs. If I re-map every three years, and I do summer and winter mapping, over a 10-year period, I will be mapping 8 times. My system cost over the same time period will be $120k. That works out to $15k per mapping session. Does that sound like a bargain? To get a mapping that is guaranteed to capture the hottest part of the summer or the coldest part of the winter? What if the sensor price and calibration costs dropped, say to $500 per sensor and $100 to recalibrate. Now our 10-year cost is $60k. Each mapping session is now $7.5k. Does it sound like a bargain now? So now the math is simple. At what sensor cost (and recalibration price) does it save money to leave the system in place for continuous mapping? A difficulty also comes in selling the idea, as the limitation may likely be the upfront cost for all the sensors. Monitoring systems are usually considered “Capital Expenses,” whereas validation and recalibration costs are usually considered “Operating Expenses.” So you really need a situation where management is focused on total cost of ownership, not just Capital Expense represented by the acquisition cost. As long as sensors and recalibration decrease in price, “Continual Mapping” will eventually make sense, especially in areas with high labor prices, or frequent remapping. But, let’s not forget new technology! Disposable loggers used for shipping purposes are currently available for $40 USD or less. They often have a +/-0.5C accuracy and a battery life that is good enough for at least 1 month of mapping. Now my 50-point mapping of the warehouse has a hardware cost of only $2000. I think these disposable devices will become the standard for small chamber mapping – refrigerators, freezers (except ultra-lows), and incubators. I also think continuous mapping will eventually make sense in warehouses. We will probably be sticking to some level of traditional mapping when we discuss areas where humidity is concerned, or when we deal with extreme temperature or pressure (lyophilizers, autoclaves, ultra-low freezers). I hope this answer was useful, and that you will send us any future questions you might have! I am speaking at both Interphex NYC and the Measurement Science Conference in Long Beach in March 2014. I’ll be talking about temperature mapping. If you have a chance, please stop by and say hi! Best regards, Paul
