Recommendations for Your Next Mapping Study: Total Times, Device Accuracy, and Must-Read Regulatory Docs Paul Daniel, Vaisala Senior GxP Regulatory Compliance Expert Published: Jun 2, 2014 Life Science Over the past couple of years, our Sr. Regulatory Expert Paul Daniel answered numerous questions by email. In this week's blog, we share the 7 most common questions on mapping/validation studies...
What type of sensor or logging device is recommended? This is really a wide open question. Many companies manufacture appropriate devices for this function. Here I have to recommend that you contact our application engineers to find the instruments and devices that would be suitable for your application. ☺ What is the required accuracy of the sensors used in a mapping study? For mapping storage, you would want a sensor that was ± 0.5°C, or better… preferably better. I personally prefer a sensor that is ±0.2°C to prevent my data from being questioned by an inspector/auditor. What is the accepted calibration frequency of sensors used for mapping studies? The sensors should be verified to be accurate before and after the mapping study. Actual calibrations may not need to be performed that often. Most devices in this class require calibration once or twice a year. In the case of a fridge or a freezer, is mapping advised in loaded or in unloaded condition? Mapping should be performed for both loaded and unloaded conditions. We have blogs with more in-depth discussions: "Chamber Mapping: Loaded, Unloaded, Percentage" "Temperature Mapping Worst-Case Scenarios in Refrigerators" What is the required sampling frequency during the mapping procedure (e.g. every 5 minutes per sensor)? There is no required data sampling frequency. The frequency should be appropriate to the storage area, however, you would be fine with a sampling frequency between 5 and 15 minutes How long does the total measuring time need to be (e.g. is 3 hours sufficient)? The total mapping time will again depend on the storage area, but here are my usual guidelines. For a chamber, such as a free-standing refrigerator, freezer, or incubator, I would like to see a minimum of 48-hours mapping. For a warehouse, I would like to see 7-days. Find more information on page 6 of this white paper: Validation/Mapping: Troubleshooting Tips and Best Practices "Your study should be long enough to provide confidence that you have accurately captured the environmental dynamics of the space being mapped. Forty-eight hours is sufficient for most small spaces under 2 m3 and 5 to 7 days is common for warehouses. However, the larger the space, and the more actively it is used during the study, the longer the expected duration of mapping. For a warehouse in use 5 days a week, a study duration of 1 week may be appropriate. Consider seasonal changes as well for large spaces, performing your mapping studies during the hottest and coldest times of the year. For all of these questions, it’s important that you develop a clear rationale for your choices and document them in your validation protocol. Your rationale should be scientifically based, appropriate to your facility and product, and suitable for the intended use of the space being mapped." Is there a generally accepted guidance that covers all these items? There is not just one guidance here. But for starters, see the following (the first 2 are for purchase): ISPE Good Practice Guide: Cold Chain Management PDA Technical Report 58: Risk Management for Temperature Controlled Distribution USP Chapter <1079> Good Storage and Distribution Practices for Drug Products I hope you find my answers useful! If you have any further questions on the topic, please don’t hesitate to reach out to me or contact our sales engineers.
What type of sensor or logging device is recommended? This is really a wide open question. Many companies manufacture appropriate devices for this function. Here I have to recommend that you contact our application engineers to find the instruments and devices that would be suitable for your application. ☺
What is the required accuracy of the sensors used in a mapping study? For mapping storage, you would want a sensor that was ± 0.5°C, or better… preferably better. I personally prefer a sensor that is ±0.2°C to prevent my data from being questioned by an inspector/auditor.
What is the accepted calibration frequency of sensors used for mapping studies? The sensors should be verified to be accurate before and after the mapping study. Actual calibrations may not need to be performed that often. Most devices in this class require calibration once or twice a year.
In the case of a fridge or a freezer, is mapping advised in loaded or in unloaded condition? Mapping should be performed for both loaded and unloaded conditions. We have blogs with more in-depth discussions: "Chamber Mapping: Loaded, Unloaded, Percentage" "Temperature Mapping Worst-Case Scenarios in Refrigerators"
What is the required sampling frequency during the mapping procedure (e.g. every 5 minutes per sensor)? There is no required data sampling frequency. The frequency should be appropriate to the storage area, however, you would be fine with a sampling frequency between 5 and 15 minutes
How long does the total measuring time need to be (e.g. is 3 hours sufficient)? The total mapping time will again depend on the storage area, but here are my usual guidelines. For a chamber, such as a free-standing refrigerator, freezer, or incubator, I would like to see a minimum of 48-hours mapping. For a warehouse, I would like to see 7-days. Find more information on page 6 of this white paper: Validation/Mapping: Troubleshooting Tips and Best Practices "Your study should be long enough to provide confidence that you have accurately captured the environmental dynamics of the space being mapped. Forty-eight hours is sufficient for most small spaces under 2 m3 and 5 to 7 days is common for warehouses. However, the larger the space, and the more actively it is used during the study, the longer the expected duration of mapping. For a warehouse in use 5 days a week, a study duration of 1 week may be appropriate. Consider seasonal changes as well for large spaces, performing your mapping studies during the hottest and coldest times of the year. For all of these questions, it’s important that you develop a clear rationale for your choices and document them in your validation protocol. Your rationale should be scientifically based, appropriate to your facility and product, and suitable for the intended use of the space being mapped."
Is there a generally accepted guidance that covers all these items? There is not just one guidance here. But for starters, see the following (the first 2 are for purchase): ISPE Good Practice Guide: Cold Chain Management PDA Technical Report 58: Risk Management for Temperature Controlled Distribution USP Chapter <1079> Good Storage and Distribution Practices for Drug Products I hope you find my answers useful! If you have any further questions on the topic, please don’t hesitate to reach out to me or contact our sales engineers.
Mario Rodríguez Feb 24, 2023 I've been searching (without success) for the minimum time the sensor needs to read each parameter and show a valid data at its output. It refers to the speed in which the sensor can perceive a change (in gases or in particulate matter) and have a readable data in its output. What is that "minimum reading time"? Thanks in advance. Reply
Paula Daniel Apr 11, 2023 Hi Mario! Thanks for participating in our blog. The value you are searching for is called “response time”. There are two parts to this. The first part is the mechanics or electronics of the sensor itself – what is the lag time necessary to get any reading, convert it from analog to digital, store it local memory (if attached to a data logger), and then send or display the data. Usually, this is so fast it is negligible, but will have added time (measured in fractions of a second) if the device is in a sleeping state to save battery life, then must wake up to take a reading. The second part is the physics of the measurement at the sensor element itself. In this case we are talking about temperature in this blog. How fast a bare sensor element can respond in controlled conditions is a known value for our products, but we don’t share it because it is meaningless (and misleading) in application. For instance, the bare sensor is usually placed in a metal sleeve, or coated with water-proofing. Then it is placed on the data logger, sometimes inside the logger, sometimes on a probe. And then the protected sensor is then placed in the actual measurement environment, where there are a ton of variables that will affect response times, such as airflow, temperature differential, rate of temperature change, mounting location, etc. In your question though, you asked specifically about gases and particles. Those work quite differently than temperature sensors, so I’ll discuss them separately. For gases concentrations, we can just talk about humidity, because that’s what is being measured – the concentration of gaseous H2O (also known as water vapor). A bare humidity sensor with have a very quick response time. But you never want your humidity sensor unprotected. It is usually behind a filter, made of plastic or metal, which can further slow the response. And then the placement of the sensor in or out of airflow will also affect the ability to detect changes in an actively controlled environment. And if you are measuring relative humidity, the measurement is relative to temperature, so the physics of temperature change (as discussed above) must also be considered. For other gases, I am not personally familiar with the sensors we use at Vaisala, but I should work similarly. For particles, the response time for a valid measurement isn’t based on the sensor (though these sensors are very fast electronically as they depend on light, often lasers). Usually a particle count measurement is presented as the number of particles of a given size per volume of air, say per cubic foot or per cubic meter. In order to achieve a reportable or meaningful number, the particle counter must process the required volume of air, and this takes time for the vacuum pump to pull the air volume through the particle counter. In the USA, the standard is per cubic foot, and the devices are usually designed to sample one cubic foot of air per minute, and one cubic meter of air every 35 min. I hope this helps. Let me know if you have any further questions. Best Regards, Paul Daniel Sr. GxP Regulatory Expert Reply