Personalized Z-Score Framework for Attendance Anomaly Detection in GPS-Based Laboratory Workforce Systems

Riadi Marta Dinata; Ashari Abidin; Marhaeni Marhaeni; Niko Purnomo

doi:10.30998/faktorexacta.v19i2.1610

Authors

Riadi Marta Dinata Author
Ashari Abidin Author
Marhaeni Marhaeni Author
Niko Purnomo Nusamandiri University Author

DOI:

https://doi.org/10.30998/faktorexacta.v19i2.1610

Keywords:

Anomaly Detection, GPS-based attendance, Personalized Baseline, Workforce Analytics, Z-score

Abstract

In the context of GPS-based attendance analytics for higher education, accurately identifying anomalous individual lateness patterns among laboratory assistants remains a critical challenge because attendance systems typically rely on rigid global thresholds that can introduce bias. This research aims to enhance fairness and interpretability in anomaly detection for human workforce attendance by proposing the Personalized Z-Score Attendance Anomaly Detection Framework (PZ-AD), which compares each laboratory assistant against his or her own historical lateness profile. The approach computes per-assistant means and standard deviations of lateness and standardizes each event into a Z-score, which is evaluated at multiple thresholds to distinguish normal events, medium anomalies, and strong anomalies. PZ-AD is applied to approximately 4,300 GPS attendance records from 13 laboratory assistants over a sixteen-month period. The quantitative results show that detected anomalies decrease sharply as the threshold increases: 449 anomalies affecting all 13 assistants were identified at 1.8, 294 anomalies affecting all 13 assistants at 2.2, and only 6 anomalies involving 2 assistants at 3.2; no anomalies were detected at 5.0. All detected anomalies occurred on the late-arrival side. These findings indicate that PZ-AD identifies deviations relative to each assistant's personal attendance behavior while making the sensitivity-selectivity trade-off explicit for managerial follow-up. The framework provides an interpretable baseline for workforce attendance analytics and establishes a foundation for future comparison with robust Modified Z-score methods for skewed lateness distributions

Downloads

Download data is not yet available.

References

[1] A. Blázquez-García, A. Conde, U. Mori, dan J. A. Lozano, “A review on outlier/anomaly detection in time series data,” ACM Computing Surveys, vol. 54, no. 3, pp. 1–33, 2021, doi: 10.1145/3444690.

[2] A. A. Cook, G. Misirli, dan Z. Fan, “Anomaly detection for IoT time-series data: A survey,” IEEE Internet of Things Journal, vol. 7, no. 7, pp. 6481–6494, 2020, doi: 10.1109/JIOT.2019.2958185.

[3] L. Correia, J. C. Goos, P. Klein, T. Bäck, dan A. V. Kononova, “Online model-based anomaly detection in multivariate time series: Taxonomy, survey, research challenges and future directions,” Engineering Applications of Artificial Intelligence, vol. 138, Art. no. 109323, 2024, doi: 10.1016/j.engappai.2024.109323.

[4] Z. Z. Darban, G. I. Webb, S. Pan, C. C. Aggarwal, dan M. Salehi, “Deep learning for time series anomaly detection: A survey,” ACM Computing Surveys, vol. 56, no. 12, pp. 1–42, 2024, doi: 10.1145/3691338.

[5] B. Iglewicz dan D. C. Hoaglin, How to Detect and Handle Outliers. Milwaukee, WI, USA: ASQC Quality Press, 1993.

[6] S. Kumar et al., “GPS based attendance system using geofencing,” International Journal of Scientific Research in Science and Technology, vol. 10, no. 3, pp. 204–211, 2023.

[7] S. Kumari et al., “A comprehensive investigation of anomaly detection methods in deep learning and machine learning: 2019–2023,” IET Information Security, Art. no. 8821891, 2024, doi: 10.1049/2024/8821891.

[8] C. Leys, M. Delacre, Y. L. Mora, D. Lakens, dan C. Ley, “How to classify, detect, and manage univariate and multivariate outliers, with emphasis on pre-registration,” International Review of Social Psychology, vol. 32, no. 1, 2019, doi: 10.5334/irsp.289.

[9] A. Pawar, R. Hiwanj, P. Koparde, dan S. Barve, “Automated employee attendance monitoring using liveness face recognition and geofencing in real time,” in Proc. Int. Conf. Communication, Computing and Internet of Things (IC3IoT), 2023, pp. 1–6.

[10] P. Patil, S. Ringe, P. Gaikwad, P. Mahajan, dan S. Mogare, “An Android application for attendance using geofencing,” International Journal for Research in Applied Science and Engineering Technology, vol. 11, no. 11, pp. 1505–1510, 2023, doi: 10.22214/ijraset.2023.56620.

[11] D. M. Ravid, D. L. Tomczak, J. C. White, dan T. S. Behrend, “The impact of electronic monitoring on employees’ job satisfaction, stress, performance, and counterproductive work behavior: A meta-analysis,” Computers in Human Behavior, vol. 127, Art. no. 107023, 2022, doi: 10.1016/j.chb.2021.107023.

[12] D. M. Ravid dan D. Jeske, “Electronic monitoring at work,” Annual Review of Organizational Psychology and Organizational Behavior, vol. 12, 2025, doi: 10.1146/annurev-orgpsych-110622-060758.

[13] P. J. Rousseeuw dan C. Croux, “Alternatives to the median absolute deviation,” Journal of the American Statistical Association, vol. 88, no. 424, pp. 1273–1283, 1993.

[14] K. R. Shahapure dan C. Nicholas, “Cluster quality analysis using silhouette score,” in Proc. IEEE 7th Int. Conf. Data Science and Advanced Analytics (DSAA), 2020, pp. 747–748, doi: 10.1109/DSAA49011.2020.00096.

[15] A. S. Yaro, F. Maly, dan P. Prazak, “Outlier detection in time-series receive signal strength observation using Z-score method with Sn scale estimator for indoor localization,” Applied Sciences, vol. 13, no. 6, Art. no. 3900, 2023, doi: 10.3390/app13063900.

[16] A. S. Yaro, F. Maly, dan P. Prazak, “Outlier detection performance of a modified Z-score method in time-series RSS observation with hybrid scale estimators,” IEEE Access, vol. 12, pp. 12345–12358, 2024, doi: 10.1109/ACCESS.2024.3356731.

[17] J. Gomez, M. Rahman, dan P. Singh, “Anomaly detection in time-series employee absence data,” in Proc. IEEE Int. Conf. Big Data (BigData), 2023, pp. 1–8.

[18] R. Khan dan S. Patel, “Geolocation and geofencing-based attendance systems for workforce management,” Journal of Workforce Analytics, vol. 7, no. 2, pp. 45–60, 2025.

[19] F. Osei, K. Mensah, dan D. Boateng, “A multi-method anomaly detection framework for identifying potential ghost employees in payroll data,” International Journal of Research in Engineering and Science, vol. 13, no. 4, pp. 1–12, 2025.

[20] R. Guidotti, A. Monreale, S. Ruggieri, F. Turini, F. Giannotti, and D. Pedreschi, "A Survey of Methods for Explaining Black Box Models," ACM Computing Surveys, vol. 51, no. 5, pp. 1-42, 2018, doi: 10.1145/3236009.

[21] A. Barredo Arrieta, N. Díaz-Rodríguez, J. Del Ser, A. Bennetot, S. Tabik, A. Barbado, S. Garcia, S. Gil-Lopez, D. Molina, R. Benjamins, R. Chatila, and F. Herrera, "Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI," Information Fusion, vol. 58, pp. 82-115, 2020, doi: 10.1016/j.inffus.2019.12.012.

[22] P. Tambe, P. Cappelli, and V. Yakubovich, "Artificial Intelligence in Human Resources Management: Challenges and a Path Forward," California Management Review, vol. 61, no. 4, pp. 15-42, 2019, doi: 10.1177/0008125619867910.

[23] Z. Li, Y. Zhu, and M. van Leeuwen, “A Survey on Explainable Anomaly Detection,” ACM Transactions on Knowledge Discovery from Data, vol. 18, no. 1, Art. no. 23, pp. 1–54, 2024, doi: 10.1145/3609333.

[24] L. Antwarg Friedman, C. Galed, L. Rokach, and B. Shapira, “Evaluating Anomaly Explanations Using Ground Truth,” AI, vol. 5, no. 4, pp. 2375–2392, 2024, doi: 10.3390/ai5040117.

[25] A. Heidemann, S. M. Hülter, and M. Tekieli, “Machine learning with real-world HR data: mitigating the trade-off between predictive performance and transparency,” The International Journal of Human Resource Management, vol. 35, no. 14, pp. 2343–2366, 2024, doi: 10.1080/09585192.2024.2335515.

[26] S. Dadaboyev, S. Hojimurodova, and G. Gulomova, “AI in HR Analytics: A Synthesis of Current Research and Future Directions,” Proceedings of the AAAI Symposium Series, vol. 6, no. 1, p. 60, 2025, doi: 10.1609/aaaiss.v6i1.36026.