The remaining SL of a product is directly related to the rate of its deterioration, which depends on external parameters such as temperature, humidity etc. Predicting 24/7 the time in which the acceptability limit is reached requires the permanent combination of these external parameters with the kinetic parameters of the deterioration of materials.

The tools currently available on the market which are used for determining the remaining shelf life are based on oversimplified kinetic approaches and are very imprecise. The situation is even worse when the temperatures fluctuate during storage or transport, which is a common issue in the last leg of the supply chain. There is a need of predictive tools to calculate the remaining shelf life of the materials at any storage place or transport time.

The potential markets for the USLM are wide-ranging, including the pharmaceutical (vaccines, drugs), food and chemical industries, the safety for the surveillance of dangerous goods including the high energetic materials in defence and space applications. All this information transmitted to the end user will enable the supplier to monitor the precise state of degradation of the materials remotely. The cost of the access to the USLM system, will be offset by the savings from avoiding product waste and preventing accidents.

In reality, a correct decision on whether a material has reached its shelf life based on temperature data only is not sufficient. The sole knowledge of temperature history is only one prerequisite.

The second, very important prerequisite is the precise determination of the kinetic parameters of the material deterioration which requires the application of advanced kinetic and statistical approaches. The knowledge of kinetic parameters allows for a very accurate description of the reaction rate and the degree of the material deterioration for any temperature profile by applying the continuously measured data under real environmental conditions.

For the transport and storage of dangerous goods that may decompose exothermally, the knowledge of both, the surrounding temperature and kinetic parameters is not sufficient. Kinetic parameters are the same for samples with mass of 1mg, 1g, 1kg or 1 ton. However, for larger mass samples, a third, also very important prerequisite, the precise heat balance in the system, should be considered. In such cases, the evolved heat due to the exothermic decomposition may not be fully exchanged with the environment, which can result in a temperature increase in the system and the thermal runaway.

Migration of the package components into the food is also a global concern. Here, in addition to the temperature history, the knowledge of diffusion and partition coefficients represents a fourth prerequisite for evaluating the rate at which organic migrants in multi-layer plastic packaging diffuse into packed items.

By accurately predicting the deterioration of materials, the USLM system will help to reduce waste. In specific markets, our aim is to extend the life of materials by accurately tracking their deterioration in real time. This is crucial because it can significantly extend expiry dates, particularly for temperature-sensitive products such as vaccines. It also makes it possible to maintain the quality of products that are less affected by temperature, such as medicines, foodstuffs and propellants. This approach is in line with Objective 12 of the 2030 Agenda for Sustainable Development, which focuses on sustainable consumption and production.

Some energetic materials change their thermal behaviour and, in turn, the safety parameters during storage. In the case of autocatalytic reactions, just 1% of the material’s degradation can lead to a drop in SADT and TMRad of around 25°C, which can be very dangerous. In the case of autocatalytic reactions, i.e. materials that continuously change their properties during storage, their temperature during storage must be carefully controlled and, in turn, the degree of ageing and the variation in the TMRad and SADT safety parameters must be monitored. The USLM system enables real-time monitoring. By integrating material decomposition kinetics and temperature data from data loggers into our USLM system, it can continuously monitor any changes in the thermal behavior of energetic materials in real time. This helps prevent runaway reactions and potential explosions. As a result, our solution will improve the assessment of the thermal safety of chemicals and help to better comply with the recommendations of the United Nations Manual of Tests and Criteria on the Transport of Dangerous Goods.

[1] Roduit B.; Hartmann M.; Folly P.; Sarbach A.; Chem. Ing. Trans., 2013, 31, 907.

[2] Recommendations on the transport of dangerous goods, Model Regulations, Vol.I, 21-th revised edition, United Nations, 2019, (accessed on 27th Dec. 2023),

https://unece.org/fileadmin/DAM/trans/danger/publi/unrec/rev21/ST-SG-AC10-1r21e_Vol1_WEB.pdf

[3] Recommendations on the transport of dangerous goods, Model Regulations, Vol.II, 21-th revised edition, United Nations, 2019, 2019, (accessed on 27th Dec. 2023),

https://unece.org/fileadmin/DAM/trans/danger/publi/unrec/rev21/ST-SG-AC10-1r21e_Vol2_WEB.pdf

Improving supply chain management through kinetic modelling using our USLM system, can offer enormous potential for solving crucial problems:

Firstly, it could significantly reduce financial losses due to cold chain breaks, a problem identified by institutions such as the World Health Organization (WHO) and the Centre for Disease Control (CDC). For example, WHO reports indicate that up to 25% of vaccines worldwide arrive at their destination in a degraded state due to temperature-related problems during transport. This problem is not confined to developing countries; it occurs wherever there is monitoring of vaccine temperature. Therefore, the USLM advances could have a profound impact on global health, particularly for vulnerable populations. In the past, more than 50% of vaccine batches worldwide have been discarded, illustrating the urgent need for better management. By optimizing the ‘cold chain’ process, which accounts for around 80% of the costs of vaccination programs, kinetic modelling in combination with our USLM system could significantly reduce these losses in supply chain management.

And the benefits are not limited to healthcare. The United Nations Food and Agriculture Organization (FAO) estimates that around a third of the world’s food production is lost or wasted each year. Improving the management of the shelf life of stored and transported food products through better supply chain practices thank to our USLM system could significantly reduce this wastage, with considerable positive repercussions for everyone reducing inequalities.

Our project focuses primarily on T-RH dataloggers produced in EU countries to meet EU standards in terms of environmental responsiveness, with the added benefit of long battery life to minimize the waste generated by our solution based on dataloggers which can potentially become very large in number. In addition, we will be focusing our development efforts on dataloggers with replaceable batteries. By allowing users to replace the batteries, we are significantly extending the life of the dataloggers, resulting in a further reduction in waste.

1.1 Highly accurate determination of (i) deterioration kinetics, (ii) thermal safety parameters or (iii) migration modeling;

1.2 Description of reaction kinetics based on conventional thermoanalytical data (e.g. DSC, etc.) but also based on limited amount of (sparse) data. Evaluation of the kinetic parameters from sparse data obtained through accelerated thermal aging should be based on modified kinetic analysis and model selection approaches allowing to:

• Screen the kinetic models
• Rank the kinetic models according to statistical analysis (AIC/BIC)
• Predict long-term stability and determine the remaining shelf life with adjustable probability for the Prediction Bands (PB with e.g. Confidence Interval CI 95%) obtained by e.g. bootstrap techniques.

1.3 Incorporation of temperature (but also humidity) influence into kinetic models;

1.4 Continuous online improvement of the parameters of the kinetic models by incorporating new data from follow-up measurements of samples exposed to any temperature variations recorded by the dataloggers;

1.5 Determination of the Self-Accelerating Decomposition Temperature (SADT);

1.6 Detection of possible autocatalytic behavior of analyzed reactions due to the permanent influence of the reaction progress on the thermal behavior of this range of materials (e.g., SADT reduction as a function of aging;

1.7 Possibility of determining the migration of a substance in packed items for comparison with Specific Migration Limit (SML);

1.8 Exporting all kinetic parameters necessary for online remaining SL, SADT and SML determination into the SLM system.

2.1 Are IoT-based with sufficient open-source resources for possible direct integration via, for example, GSM/LTE into the USLM system, or for the transmission of the collected data over, for example, Bluetooth Low Energy (BLE) to gateways;

2.2 Have a small size to be inserted into existing packaging (reusability).

2.3 Have sufficient measurement accuracy with

• broad temperature range of application from -40°C to +85°C
• possibility to measure temperature and humidity, as well as other interesting experimental data such as atmospheric pressure, acceleration (in g) in the x, y and z directions to determine the position of the package, and even be able to count the number of movements
• battery life (at least 1-2 years of autonomy), emission, etc.;

2.4 Have a statement indicating that they meet International Certification Standards;

2.5 Offer straightforward installation;

2.6 Offer the possibility to secure data transmission;

2.7 Are inexpensive (around ~ 25 EUR including battery).

3.1 Communicate wirelessly with dataloggers, for example, over BLE and collect temperature and humidity data;

3.2 Send the temperature and humidity data from any location into the USLM system;

3.3 Provide sufficient open-source features to integrate the collected temperature and humidity data directly into the USLM system, allowing for the building of a remote monitoring system that works with all wireless T-RH dataloggers in proximity, regardless of whether there are only a few or hundreds of them;

3.4 Allow for easy and fast installation;

3.5 Offer the possibility to secure data transmission.

4.1 Allow for seamless sending of GPS coordinates over networks for direct integration into the USLM system;

4.2 Be able to collect and send temperature and humidity data (acting as a gateway) in addition to latitude and longitude (all in one solution);

4.3 Facilitate fast installation of GPS and, if BLE is available, easy connection to BLE dataloggers;

4.4 Offer the possibility to secure data transmission.

5.1 Online servers and cloud technology are utilized to:

• send, receive, manage, and store data, as well as transfer the data to a web portal platform or mobile applications;
• perform real-time remote monitoring of objects that require 24/7 supervision;

5.2 The cold chain is maintained by ensuring:

5.3 Online servers and cloud technology are utilized to:

• send, receive, manage, and store data, and transfer it to a web portal platform or mobile applications;
• perform real-time remote monitoring of products that require 24/7 supervision, including:

– the state of material aging;
– remaining shelf life time;
– alarm, warning, and supervision management via emails and SMS.

5.4 The cold chain is maintained by ensuring:

• constant calculation of the current extent of product deterioration;
• continuous evaluation of the remaining shelf life time at the current temperature or any new temperature profile;
• automatic alarm notifications via email/SMS if, for example, the remaining shelf life time is zero.

5.5 Thermal safety is maintained through:

• continuous surveillance during the transport and storage of dangerous goods;
• continuous evaluation of possible changes in Self-Accelerating Decomposition Temperature (SADT) as a function of thermal aging;
• Automatic alarm notifications via email/SMS if, for example, T-SADT is less than 10°C.

5.6 Food safety is ensured through:

• continuous evaluation of migration into packed items;
• automatic alarm notifications via email/SMS if, for example, migration exceeds the Specific Migration Limit (SML).

5.7 Cloud computing is used for various predictive analyses.

The information received from dataloggers, vaccine vial monitors, peak temperature threshold indicators, and other devices recommended by the WHO represent the state of the art and are the only indicators as to whether the storage fulfills a specific refrigeration temperature criterion. The recent implementation of continuous evaluation of the so-called Mean Kinetic Temperature (MKT) into dataloggers does not change the situation in which it is impossible to continuously monitor the actual degree of the deterioration of samples. The MKT is essentially just another way to express the impact of temperature during sample exposure which does not bring information of whether the permissible reaction progress limit is reached. The MKT concept has also significant drawbacks, as it assumes that deterioration is a one-step first order reaction for all materials, with a fixed activation energy (83.144 kJ·mol−1 for all decomposing materials!) and a fixed time interval for data collection. A detailed description of the MKT concept can be found in [1], its modification in [2], and limitations in [3], where one finds the recommended caution in using the MKT to evaluate temperature excursions. It was also presented, that the sole focus on the activation energy, even if the SADT is the same, may result in accidents if a e.g. a first order model is assumed for a deterioration that is actually strongly autocatalytic [4]. It is nowadays known that the deterioration of most materials, chemicals or biopharmaceuticals is linked to multistage overlapped reactions with more than one activation energy and different reaction models.

In summary, a lot of datalogger solutions exist today and some of them even propose a basic shelf life calculation, but to our knowledge no solution is as advanced as AKTS calculation methods. The USLM system’s advanced kinetic-based approach through the AKTS-software suite is by far more advanced and will therefore lead to substantial economic and ecological savings compared to international state of the art.

[1] Seevers, R.H.; Hoffer, J.; Harber, P.; Ulrich, D.A.; Bishara, R. The use of Mean Kinetic temperature (MKT) in the handling, storage and distribution of temperature sensitive pharmaceuticals. Pharm. Outsourcing 2009, 12–17, Available online: https://studylib.net/doc/8854836/the-use-of-mean-kinetic-temperature–mkt-, (Accessed on: 20 June 2024).

[2] Tong, C.; Lock, A. A computational procedure for Mean Kinetic Temperature using unequally spaced data. In Proceedings of the Joint Statistical Meeting 2015, Biopharmaceutical Section, Seattle, WA, USA, 8–13 August 2015; pp. 2065–2070.

[3] Roduit B.; Luyet C.-A.; Hartmann M.; Folly P.; Sarbach A.; Dejeaifve A.; Dobson R.; Schroeter N.; Vorlet O.; Dabros M.; Baltensperger R.; Molecules, 2019, 24, 2216.

[4] Roduit B.; Hartmann M.; Folly P.; Sarbach A.; Chem. Ing. Trans., 2013, 31, 907.

The strategy consists in customizing and selling commercially available data loggers and data collectors, with a small mark-up to keep prices in line with our customers’ expectations. Our approach is to supply one data collector for every 50 data loggers to be stored within a company. These collectors are placed in specific rooms, close to the loggers, to collect data from the surrounding loggers.

For transport, we intend to supply sets of collectors linked to 1 to 5 data loggers. These collectors are fitted with a SIM card for online access and transmission of the data collected. Their autonomy varies from a few months to several years, depending on various factors such as the brand. Some manufacturers have stepped up their research to reduce power consumption and thus increase the autonomy. The type and frequency of measurements also play a crucial role. A solution that collects data every 5 seconds with the GPS position and transmits data every minute will consume more energy than a solution without GPS, collecting data every hour without transmitting the GPS position. By putting the collectors on standby when not in use, their autonomy can be extended from a few months to several years. Even in particularly demanding conditions, the integration of a small solar panel (the size of an A4 sheet of paper) to recharge the battery of the collectors on a daily basis ensures unlimited autonomy. This configuration is ideal for isolated regions with no electricity supply.

The choice of collectors and their brand therefore depends on a number of factors, including

• The specific type of storage or transport planned
• The autonomy required for the collector’s operation,
• The type, frequency and mode of data transmission,
• The need for a geolocation function.

These criteria determine the optimum selection of collectors for each situation and ensure maximum performance from data collection devices, whether they are dedicated to storage or transport.

The energy source in SLC should have a lifetime of at least 2 years. To reach this objective, it is necessary to apply a very low power consumption and small size electronics. The sensors must ensure a very precise measurement (of the order of 0.1°C for the temperature), which is a necessity for the accurate calculation of the kinetics. Additionally, it must have a flash memory to store all the data needed for the recalculation of kinetics. The battery will have to have sufficient capacity, small dimensions and be able to withstand the current peaks necessary to store the data in the flash memory as well as the wireless transmission of the data. The battery must also perform well in a wide temperature range (-40°C to +85°C). Experiments (on a short time scale and over all temperature ranges) on different types of batteries will make it possible to make a technological choice to guarantee the 2 year objective.

Once the choice of the battery is made, it will be a question of managing the storage, the transfer and the kinetic calculations in an optimal way. It will be necessary to optimize the following stages:

1) Storage: structure data to limit the energy-consuming access to the flash memory

2) Data transmission: limit, as much as possible, the transmission of information while ensuring real-time monitoring

3) Calculation of kinetics: minimizing calculations while ensuring the possibility of adapting and recalculating the kinetics during the process. In that sense, we have to develop adapted mathematical formulas to be integrated also in the mobile phone in order to decrease the power consumption.

Combining all above mentioned tasks (ultra-low power consumption electronics, advanced kinetic analysis, mathematical optimization and statistics) is a major challenge for this project. Finally, our invention will result in high-precision monitoring of material state in real-time which is non-invasive and works in extended period of time.

(1) The combination of the

• novel kinetic analysis workflow
• critical parameters
• online monitoring of the temperature profile

enables transport and storage of materials to be optimized by

•  continuously and remotely comparing the critical time with the required time, i.e. by computing the so called remaining shelflife time during storage and transport
• preventing unnecessary material destruction, runaway scenario and food stuff contamination.

(2) Our entire system is modular:

• it can be strictly reduced to temperature data only
• it can be connected to existing BLE data loggers
• it will always propose the best kinetic model that best describes the data.

1) Electronic part of datalogger

• Search for long-life battery and electronics with very low power consumption which are adapted to work in low temperatures (-40°C)
• Achieve Service life test according to operating temperature ranges

2) Develop the software and evtl customize the firmware for acquisition, storage, kinetics calculation, and data transmission

3) Data processing and visualization, control and configuration of kinetic parameters in the USLM system on both computer and mobile phone interfacing with thermokinetics Software. Development of the smart cloud system.

4) Gateway to dataloggers and communication with the smart cloud system: Data Collector Device

• Implementation of the data collector device
• Development of the mobile application

1) Development of advanced kinetics models for the prediction of the degradation of materials (integration of different parameters in the reaction progress and expansion of the kinetic models used).

2) Development of an advanced model selection approach (model averaging, Bayesian statistics & kinetics, model and parameter validation…) for sparse data and scale-up using various numerical techniques such as finite differences, finite volume approximations or finite element analysis. The final goal is to develop an ideal scenario in order to select the best predictive and confident (giving precise PB and CB) model.

3) Integration of the advanced kinetic models and advanced model selection at the mobile phone level. For this part, a particular attention will be placed in the complexity of the models used and implemented into the USLM system.

4) General goal: precise monitoring and prediction of material degradation and low power consumption at the datalogger level.

1) The dataloggers shall operate for a minimum of 2 years, ensuring real-time (every 15 minutes) monitoring of material degradation.

2) Integration of kinetics into the USLM system via microservices and mobile phone via app for computing the remaining shelf life time.

3) Bi-directional data transmission from the Thermokinetics software to the datalogger via the collectors and mobile phones.

4) Automatic selection of the kinetic model and determination of the prediction bands are performed using advanced statistical tools developed in the AKTS-Thermokinetics software. The tools will deliver the best kinetic model with e.g. 95% confidence for the prediction bands using the bootstrap concept widely applied in statistics but rarely combine with kinetics.

5) Universal Shelf Life Monitoring system and Web Portal

• Overview in real-time of all monitored parameters: temperature, relative humidity, luminosity.
• Change of the product properties (shelf-life and reaction progress).
• Management and assignment of deterioration kinetics depending on the datalogger missions.
• Continuous evaluation of the data transmitted by the datalogger.

• Integration of advanced predictive analysis for remaining shelf life time of materials.

The concept consists of the complete bidirectional IoT chain. Using AKTS-Thermokinetics Software in the laboratory, it is possible to determine the kinetic parameters of the degradation process of the materials. These kinetic parameters are transmitted to the USLM system to perform shelf life calculation. In turn, the data loggers used by the end-user transfer the collected data (such as the temperature, humidity, light) back to the USLM system. Recording at all time the experimental conditions, it is also possible to « backtrack » the substance to the laboratory providing this way more data for kinetic analysis. Having more data (i.e. more information about the sample degradation under varying conditions), it is possible to refine the kinetic parameters. The accuracy of the kinetic description of the samples aging can be this way improved thanks to the fine tuning of the kinetic parameters. They can be transferred again to the USLM system and used for fine-tuning of the remaining shelf life of the stored or transported goods. This mode of operation aims to offer real-time monitoring and guarantees the precise and actual evaluation of the remaining shelf life. Just before the product is used, the comparison of the aging extent of the material displayed by the USLM system or the mobile phone with its acceptable limit informs the user about the possibility of using the product or not. The additional technological and scientific advances will be located at the level of the combination of the cloud concept with the advanced kinetic software. The first challenge of the technological solution to be developed is to create a cloud capable of interacting with the main dataloggers on the market. It requires the optimization of the collection and smart communication of data. A second challenge lies in elaboration of advanced kinetics which allows to predict the state of the material at all time and estimate the remaining shelf life depending on e.g. the real temperature fluctuations during storage, transport and last-mile delivery. Combining all above-mentioned tasks (ultra-low power consumption electronics, advanced kinetic analysis, mathematical optimization and statistics) is a major challenge for this project. Finally, our invention will result in high-precision monitoring of material state in real-time which is non-invasive and works in extended period of time.

The datalogger’s power source should have a lifespan of 2 years or allow the batteries to be changed during use. To achieve this objective, it is necessary to apply electronics with very low energy consumption and small size. The sensors must be able to measure very precisely (of the order of 0.1°C for temperature), which is necessary for accurate calculation of kinetics. In addition, they must have flash memory to store all the data needed to recalculate the kinetics. Assuming a lifespan of 2 years and data sampling every 15 minutes, it will be necessary to choose dataloggers with a memory capable of storing thousands of measurements. The battery will need to have sufficient capacity, small dimensions and be able to withstand the current peaks required to store the data in the flash memory and to transmit the data wirelessly. The battery must also operate over a wide temperature range (-40°C to +85°C). Experiments (in the short term and over all temperature ranges) on different types of battery will enable a technological choice to be made to guarantee the 2-year target.

Once the choice of the battery is made, it will be a question of managing the storage, the transfer and the kinetic calculations in an optimal way. It will be necessary to optimize the following stages:

• Storage: structure data to limit the energy-consuming access to the flash memory
• Data transmission: insure, as much as possible, the transmission of information while for real-time monitoring
• Calculation of kinetics: minimizing calculations while ensuring the possibility of adapting and recalculating the kinetics during the process.

To expand the AKTS market, we intend to implement and compare the different existing kinetic approaches (frequentist model-based and even iso-conversional approaches) and design a precise mathematical elaboration of the collected data.

The aim will be to find the optimal solution using

• Kinetic analysis and optimization: depending on the available data a kinetic approach will be chosen in further analysis.
• Statistics: to select the best kinetic models, advanced criteria of model selection using the both the AIC and BIC criteria will be used or a Bayesian approach will be applied.
• Prediction bands: Based on the bootstrap sets of estimated kinetic parameters, the prediction is enhanced by the estimation of the prediction bands (PB) in the form of e.g., the upper and lower 95 percentiles (PB 95% confidence). The evaluation of PBs is necessary to trigger of the alarm when they go over the acceptable limit of the product aging degree.

1. https://www.aktscommunications.com
2. https://www.akts.com
3. https://www.akts.com/tk
4. https://www.akts.com/tksd
5. https://www.akts.com/ts
6. https://unece.org/fileadmin/DAM/trans/danger/publi/unrec/rev21/ST-SG-AC10-1r21e_Vol1_WEB.pdf (accessed on 20th Nov. 2024).
7. https://unece.org/fileadmin/DAM/trans/danger/publi/unrec/rev21/ST-SG-AC10-1r21e_Vol2_WEB.pdf (accessed on 20th Nov. 2024).
8. https://www.akts.com/sml
9. http://data.europa.eu/eli/reg/2011/10/oj (accessed on 1st Nov. 2024).
10. https://echa.europa.eu/plastic-material-food-contact (accessed on 1st Nov. 2024).
11. Roduit B.; Hartmann M.; Folly P.; Sarbach A.; Baltensperger R.; Thermochim. Acta, 2014, 579, 31.
12. Clénet D.; Imbert F.; Probeck P.; Rahman N.; Ausar S.; J. Pharma. Sci., 2014, 103, 3055.
13. Clénet D.; Europ. J. Pharm. Biopharm., 2018, 125, 76.
14. Roduit B.; Luyet C.-A.; Hartmann M.; Folly P.; Sarbach A.; Dejeaifve A.; Dobson R.; Schroeter N.; Vorlet O.; Dabros M.; Baltensperger R.; Molecules, 2019, 24, 2216.
15. Neyra C.; Clénet D.; Bright M.; Kensinger R.; Hauser S.; Int. J. Pharm., 2021, 609, 121143.
16. Evers A.; Clénet D.; Pfeiffer-Marek S.; Pharmaceutics, 2022, 14, 375.
17. Huelsmeyer M.; Kuzman D.; Bončina M.; Martinez J.; Steinbrugger C.; Clénet D.; et al.; Scientific Reports, Nature, 2023, 13,10077.
18. Roduit B.; Folly P.; Sarbach A.; Baltensperger R.; (https://europyro2023.org, accessed on 1st Nov. 2024).
19. Chau J.; Altan S.; Burggraeve A.; Coppenolle H.; Kifle Y.W.; Prokopcova H.; Van Daele T.; Sterckx H.; AAPS PharmSciTech, 2023, 24, 250.
20. Brown M. E.; Maciejewski M.; Vyazovkin S.; Nomen R.; Sempere J.; Burnham A.; et al.; Thermochim. Acta, 2000, 355, 125.
21. Roduit B.; The ICTAC Kinetics Project, Thermochim. Acta, 2000, 355, 171.
22. Roduit B.; Xia L.; Folly P., Berger B.; Mathieu J.; Sarbach A.; Andres H.; Ramin M.; Vogelsanger B.; Spitzer D.; Moulard H.; Dilhan D.; J. Therm. Anal. Cal., 2008, 93, 143.
23. Roduit B.; Dermaut W.; Lunghi, A.; Folly P.; Berger B.; Sarbach A.; J. Therm. Anal. Cal., 2008, 93 1, 163.
24. Vyazovkin S.; Burnham A.K.; Criado J.M.; Pérez-Maqueda L.A.; Popescu C.; Sbirrazzuoli N.; Thermochim. Acta, 2011, 520, 1.
25. Vyazovkin S.; Chrissafis K.; Di Lorenzo M.L.; Koga N.; Pijolat M.; Roduit B.; Sbirrazzuoli N.; Sunol J.J.; Thermochim. Acta, 2014, 590, 1.
26. Dobson R.; Folly P.; Sarbach A.; Van Riet R.; Roduit B.; et al.; Propell. Explos. Pyrotech., Apr. 2024, 49 (4).
27. Stoessel F.; Thermal Safety of Chemical Processes, Risk Assessment and Process Design, 2. Auflage – März 2020, WILEY-VCH Verlag GmbH & Co. CGaA.
28. Keller A.; Stark D.; Fierz H.; Heinzle E.; Hungerbuehler K.; Ind. J. Loss Prev. Process, 1997, 10(1),31.
29. H. Fierz; J. Hazard. Mater., 2003, A96, 121.
30. Roduit B.; Folly P.; Berger B.; Mathieu J.; Sarbach A.; Andres H.; Ramin M.; Vogelsanger B.; J. Therm. Anal. Cal., 2008, 93, 153.
31. Roduit B.; Folly P.; Brogli F.; Mascarello F.; Glarner T.; Irle E.; Wiss J.; Luginbühl M.; Stoessel F.; Chem. Propell. Polym. Mater., 2011, 9(1), 84.
32. Dellavedova M.; Pasturenzi C.; Gigante L.; Lunghi A.; Chem. Eng. Trans., 2012, 26, 585.
33. Roduit B.; Hartmann M.; Folly P.; Sarbach A.; Brodard P.; Baltensperger R.; J. Therm. Anal. Calorim., 2014, 117, 1017.
34. Roduit B.; Hartmann M.; Folly P.; Sarbach A.; Brodard P.; Baltensperger R.; Thermochim. Acta 2015, 621, 6.
35. Roduit B.; Hartmann M.; Folly P.; Sarbach A.; Dejeaifve A.; Dobson R.; Kurko K.; Thermochim. Acta, 2018, 665, 119.
36. Cong Y.-B.; Wei Z.-Y.; Ma X.-H.; Li Z.-L.; Li Q.-G.; Mong X.; Cheng C.-S.; Results Chem., 2021, 3, 100112.
37. Roduit B.; Borgeat C.; Cavin S., Fragnière C.; Dudler V.; Food Addit. Contamin. 2005, 22 (10), 945.
38. https://www.oig.hhs.gov/oei/reports/oei-04-10-00430.pdf (accessed on 1st Nov. 2024).
39. https://www.futuremarketinsights.com/reports/vaccines-market (accessed on 15th Nov. 2024).
40. https://www.precedenceresearch.com/polymers-market (accessed on 15th Nov. 2024).
41. https://www.databridgemarketresearch.com/reports/global-process-safety-services-market (accessed on 15th Nov. 2024).
42. https://www.akts.com/dsc-arc.
43. https://www.alliedmarketresearch.com/food-safety-testing-market (accessed on 15th Nov. 2024).
44. https://kep-technologies.com (accessed on 1st Nov. 2024).
45. https://setaramsolutions.com (accessed on 1st Nov. 2024).
46. https://setaramsolutions.com/software/akts-thermokinetics (accessed on 1st Nov. 2024).
47. https://www.akts.com/about-us (accessed on 1st Nov. 2024).
48. Kartoglu U.; Milstien J.; Expert Rev. Vaccines, 2014, 13(7), 843.
49. Zipursky S.; Djingarey M.H.; Lodjo J.-C.; Olodo L.; Tiendrebeogo S.; Ronveaux O.; Vaccine, 2014, 32(13), 1431.
50. https://studylib.net/doc/8854836/the-use-of-mean-kinetic-temperature–mkt- (accessed on 15th Nov. 2024).
51. https://ww2.amstat.org/meetings/proceedings/2015/data/assets/pdf/234053.pdf (accessed on 15th Nov. 2024).
52. Okeke C.C.; Bailey L.C.; Medwick T.; Grady L.T.; Pharma. Forum, May-June 1997, 23(3), 4155.
53. Okeke C.C.; Bailey L.C.; Lindauer R.F.; Medwick T.; Grady, L.T.; Pharma. Forum, Sept.-Oct.1998, 24(5), 7064.
54. https://media.tghn.org/medialibrary/2020/12/20201209_COVAX_Storage_temperature_and_shelf_life_workshop_presentation.pdf (accessed on 20th Nov. 2024).
55. https://www.coe.int/en/web/un-agenda-2030/home (accessed on 4th Nov. 2024).
56. https://www.fao.org/newsroom/detail/Cutting-food-waste-to-feed-the-world/en (accessed on 4th Nov. 2024).
57. https://iris.who.int/bitstream/handle/10665/44169/9789241563864_eng.pdf (accessed on 4th Nov. 2024).
58. https://apps.who.int/iris/bitstream/handle/10665/68463/WHO_VB_03.18.Rev.1_eng.pdf?sequence=1&isAllowed=y (accessed on 4th Nov. 2024).
59. https://www.cdc.gov/vaccines/hcp/admin/storage/index.html (accessed on 4th Nov. 2024).
60. https://pmc.ncbi.nlm.nih.gov/articles/PMC9018100/pdf/40794_2022_Article_167.pdf (accessed on 6th Nov. 2024).
61. http://apps.who.int/iris/bitstream/10665/183583/1/WHO_IVB_15.04_eng.pdf (accessed on 11th Nov. 2024).
62. Roduit B.; Hartmann M.; Folly P.; Sarbach A.; Chem. Ing. Trans., 2013, 31, 907.
63. Campa C.; Pronce T.; Paludi M.; Weusten J.; Conway L.; Savery J.; Richards C.; Clénet D.; Vaccines, 2021, 9(10), 1114.