A Multi-Output CNN1D–GRU–MLP Architecture for Thermal Time-to-Risk Classification in Cold Chain Logistics

Authors

  • Tania Ofelia López Zerón Instituto Tecnológico de Pachuca https://orcid.org/0009-0009-3446-5942
  • Adrián Sánchez Ortega Instituto Tecnológico de Pachuca https://orcid.org/0009-0005-6282-8297
  • Marcelo Quezada Quiterio Instituto Tecnológico de Pachuca
  • Enrique de Jesús Mohedano Torres Instituto Tecnológico de Pachuca
  • Uriel Marrón Hernández Instituto Tecnológico de Pachuca
  • Gabriel Jiménez Zerón Universidad Autónoma del Estado de Hidalgo https://orcid.org/0000-0001-6664-7120

DOI:

https://doi.org/10.61467/2007.1558.2027.v18i1.1465

Keywords:

Cold chain logistics, Thermal time series, CNN1D-GRU, Logística de la cadena de frío, series temporales térmicas

Abstract

Maintaining an uninterrupted cold chain is essential for ensuring the safety and quality of food and pharmaceutical products. However, conventional monitoring systems mainly provide descriptive information and offer limited support for anticipating thermal risks. This study presents a hybrid CNN1D-GRU-MLP multi-output architecture for thermal time-to-risk classification in multizone cold-chain environments. Due to the limited availability of labeled operational data, the model was initially evaluated using a synthetic dataset generated from energy-balance formulations. Using 30-minute thermal observation windows, the architecture simultaneously performs four tasks: temperature forecasting, time-to-risk interval classification, thermal state classification, and critical zone identification. The interval-bases formulation provides decision-oriented time ranges that are more useful for logistics operations than exact minute-level predictions. Although validation with real-word data remains necessary, the proposes methodology demonstrates the potential to transform thermal trajectories into meaningful information that supports earlier interventions and more effective cold-chain management.

 

Spanish-language metadata / Metadatos en español
Título en español:
Arquitectura CNN1D–GRU–MLP multisalida para la clasificación del tiempo térmico hasta el riesgo en la logística de la cadena de frío

Resumen:
Mantener una cadena de frío ininterrumpida es esencial para garantizar la seguridad y la calidad de los productos alimentarios y farmacéuticos. Sin embargo, los sistemas convencionales de monitorización proporcionan principalmente información descriptiva y ofrecen un apoyo limitado para anticipar riesgos térmicos. Este estudio presenta una arquitectura híbrida multisalida CNN1D-GRU-MLP para la clasificación del tiempo térmico hasta el riesgo en entornos multizona de cadena de frío.

Debido a la disponibilidad limitada de datos operativos etiquetados, el modelo se evaluó inicialmente utilizando un conjunto de datos sintéticos generado a partir de formulaciones de balance de energía. Mediante ventanas de observación térmica de 30 minutos, la arquitectura realiza simultáneamente cuatro tareas: predicción de temperatura, clasificación del intervalo de tiempo hasta el riesgo, clasificación del estado térmico e identificación de la zona crítica.

La formulación basada en intervalos proporciona rangos temporales orientados a la toma de decisiones que resultan más útiles para las operaciones logísticas que las predicciones exactas a nivel de minutos. Aunque sigue siendo necesaria la validación con datos reales, la metodología propuesta demuestra el potencial para transformar las trayectorias térmicas en información significativa que facilite intervenciones más tempranas y una gestión más eficaz de la cadena de frío.


Palabras Claves:
Logística de la cadena de frío, series temporales térmicas, CNN1D-GRU.

Smart citations:
SciteAI
Dimensions.
Open Alex.

References

Adefemi, K. O., Mutanga, M. B., & Alimi, O. A. (2025). A hybrid CNN–GRU deep learning model for IoT network intrusion detection. Journal of Sensor and Actuator Networks, 14(5), Article 96. https://doi.org/10.3390/jsan14050096

Artuso, P., Rossetti, A., Minetto, S., Marinetti, S., Moro, L., & Del Col, D. (2019). Dynamic modeling and thermal performance analysis of a refrigerated truck body during operation. International Journal of Refrigeration, 99, 288–299. https://doi.org/10.1016/j.ijrefrig.2018.12.014

Bavaresco, R., Arruda, H., Rocha, E., Barbosa, J., & Li, G.-P. (2021). Internet of Things and occupational well-being in Industry 4.0: A systematic mapping study and taxonomy. Computers & Industrial Engineering, 161, Article 107670. https://doi.org/10.1016/j.cie.2021.107670

Ben Taher, M. A., Kousksou, T., Zeraouli, Y., Ahachad, M., & Mahdaoui, M. (2021). Thermal performance investigation of door opening and closing processes in a refrigerated truck equipped with different phase change materials. Journal of Energy Storage, 42, Article 103097. https://doi.org/10.1016/j.est.2021.103097

Chen, K.-Y., & Shaw, Y.-C. (2011). Applying back propagation network to cold chain temperature monitoring. Advanced Engineering Informatics, 25(1), 11–22. https://doi.org/10.1016/j.aei.2010.05.003

Duan, F., Meng, X., Wu, W., Zou, Y., & Zeng, X. (2026). Data-driven virtual sensor systems for dynamic temperature monitoring along food supply chains. Journal of Stored Products Research, 115, Article 102844. https://doi.org/10.1016/j.jspr.2025.102844

U.S. Food and Drug Administration. (2023). Grade “A” Pasteurized Milk Ordinance (PMO): 2023 revision. U.S. Department of Health and Human Services. FDA official PDF

He, H., & Garcia, E. A. (2009). Learning from imbalanced data. IEEE Transactions on Knowledge and Data Engineering, 21(9), 1263–1284. https://doi.org/10.1109/TKDE.2008.239

Hoang, H. M., Akerma, M., Mellouli, N., Le Montagner, A., Leducq, D., & Delahaye, A. (2021). Development of deep learning artificial neural networks models to predict temperature and power demand variation for demand response application in cold storage. International Journal of Refrigeration, 131, 857–873. https://doi.org/10.1016/j.ijrefrig.2021.07.029

Hu, J., Sari, O., Eicher, S., & Rakotozanakajy, A. R. (2009). Determination of specific heat of milk at different fat content between 1 °C and 59 °C using micro DSC. Journal of Food Engineering, 90(3), 395–399. https://doi.org/10.1016/j.jfoodeng.2008.07.009

International Organization for Standardization. (2009). Milk—Determination of freezing point—Thermistor cryoscope method (Reference method) (ISO Standard No. 5764:2009). ISO official record

Johnson, J. M., & Khoshgoftaar, T. M. (2019). Survey on deep learning with class imbalance. Journal of Big Data, 6, Article 27. https://doi.org/10.1186/s40537-019-0192-5

Khan, S. H., Hayat, M., Bennamoun, M., Sohel, F. A., & Togneri, R. (2018). Cost-sensitive learning of deep feature representations from imbalanced data. IEEE Transactions on Neural Networks and Learning Systems, 29(8), 3573–3587. https://doi.org/10.1109/TNNLS.2017.2732482

KRESS Fahrzeugbau GmbH. (n.d.). Refrigerated truck bodies using an IVECO Daily chassis—Our CoolerBox2.0 has a lot to offer. Retrieved August 7, 2026, from KRESS Iveco Daily page

Maiorino, A., Petruzziello, F., & Aprea, C. (2021). Refrigerated transport: State of the art, technical issues, innovations and challenges for sustainability. Energies, 14(21), Article 7237. https://doi.org/10.3390/en14217237

Mengist, W., Soromessa, T., & Legese, G. (2020). Method for conducting systematic literature review and meta-analysis for environmental science research. MethodsX, 7, Article 100777. https://doi.org/10.1016/j.mex.2019.100777

Mutambik, I. (2026). A blockchain–IoT–ML framework for sustainable vaccine cold chain management in pharmaceutical supply chains. Systems, 14(5), Article 467. https://doi.org/10.3390/systems14050467

Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., Brennan, S. E., Chou, R., Glanville, J., Grimshaw, J. M., Hróbjartsson, A., Lalu, M. M., Li, T., Loder, E. W., Mayo-Wilson, E., McDonald, S., … Moher, D. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ, 372, n71. https://doi.org/10.1136/bmj.n71

Page, M. J., Moher, D., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., Brennan, S. E., Chou, R., Glanville, J., Grimshaw, J. M., Hróbjartsson, A., Lalu, M. M., Li, T., Loder, E. W., Mayo-Wilson, E., McDonald, S., … McKenzie, J. E. (2021). PRISMA 2020 explanation and elaboration: Updated guidance and exemplars for reporting systematic reviews. BMJ, 372, n160. https://doi.org/10.1136/bmj.n160

Slaghuis, B. A. (2001). The freezing point of authentic and original farm bulk tank milk in the Netherlands. International Dairy Journal, 11(3), 121–126. https://doi.org/10.1016/S0958-6946(01)00043-7

Thermo King. (n.d.). V-600. Retrieved August 7, 2026, from Thermo King V-600

United Nations Economic Commission for Europe. (2024). ATP handbook 2023: To be used with the Agreement on the International Carriage of Perishable Foodstuffs and on the Special Equipment to be Used for Such Carriage (ATP) as amended on 22 June 2024. United Nations. UNECE ATP Handbook 2023 PDF

United Nations Environment Programme, & Food and Agriculture Organization of the United Nations. (2022). Sustainable food cold chains: Opportunities, challenges and the way forward. https://doi.org/10.4060/cc0923en

Wang, Z., Dong, Y., Liu, W., & Ma, Z. (2020). A novel fault diagnosis approach for chillers based on 1-D convolutional neural network and gated recurrent unit. Sensors, 20(9), Article 2458. https://doi.org/10.3390/s20092458

Xie, Z., Long, H., Ling, C., Zhou, Y., & Luo, Y. (2025). An anomaly detection scheme for data stream in cold chain logistics. PLOS ONE, 20(3), e0315322. https://doi.org/10.1371/journal.pone.0315322

Zhang, K., Huo, X., & Shao, K. (2023). Temperature time series prediction model based on time series decomposition and Bi-LSTM network. Mathematics, 11(9), Article 2060. https://doi.org/10.3390/math11092060

Zhang, Y., Zou, Y., Liu, L., Manzardo, A., Wang, X., & Wu, J. (2026). A digital twin–driven dynamic shelf life approach for mitigating food loss across fruit supply chains. Journal of Stored Products Research, 117, Article 103016. https://doi.org/10.1016/j.jspr.2026.103016

Downloads

Published

2026-09-21

How to Cite

López Zerón, T. O., Sánchez Ortega, A., Quezada Quiterio, M., Mohedano Torres, E. de J., Marrón Hernández, U., & Jiménez Zerón, G. (2026). A Multi-Output CNN1D–GRU–MLP Architecture for Thermal Time-to-Risk Classification in Cold Chain Logistics. International Journal of Combinatorial Optimization Problems and Informatics, 18(1), 237–255. https://doi.org/10.61467/2007.1558.2027.v18i1.1465

Issue

Section

ITP 55 Anniversary Special Issue

Most read articles by the same author(s)