@inproceedings{10.1145/3774748.3787648,
author = {Puccinelli, Niccol{\`o}},
title = {Preventing Failures of Smart Human-Centric Ecosystems},
year = {2026},
isbn = {9798400722967},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3774748.3787648},
doi = {10.1145/3774748.3787648},
abstract = {Smart Human-centric Ecosystems (SHEs), like smart cities, emerge from the co-existence of heterogeneous systems with independently defined specifications. SHEs may fail despite the correct behavior of the systems that comprise the SHE, yet current approaches overlook the critical need of ensuring the reliability of SHEs. Our main goal is to ensure the reliability of SHEs by testing, predicting and ultimately preventing failures in SHEs. We propose (i) a clear and precise definition of the quality of SHEs in terms of healthiness and failures, (ii) a transformer-autoencoder anomaly detection approach to predict SHE failures, (iii) an approach to prevent SHE failures and restore the normal state of the SHE, and (iv), a multi-agent architecture for Digital Twins (DTs) of SHEs.},
booktitle = {Proceedings of the 2026 IEEE/ACM 48th International Conference on Software Engineering},
pages = {203–205},
numpages = {3},
keywords = {Smart Human-centric Ecosystems, Digital Twins, Failure Prediction, Autoencoders},
location = {
},
series = {ICSE-Companion '26}
}