A new machine learning framework for occupational accidents forecasting with safety inspections integration

2025-06-30

AI Summary

This study addresses the challenge of short-term occupational accident prediction by proposing the first binary time-series modeling framework integrating safety inspection data. Methodologically, daily accident occurrences are modeled as binary sequences; models—including logistic regression, tree-based classifiers, and LSTM—are systematically evaluated using sliding-window cross-validation and weekly-aggregated performance metrics. The LSTM achieves a balanced accuracy of 0.86, significantly outperforming baseline methods. The key contributions are: (1) the first empirical validation that safety inspection data meaningfully enhance high-risk period forecasting, and (2) the development of an end-to-end weekly risk scoring mechanism. This framework enables decision-makers to dynamically optimize resource allocation and intervention timing, thereby improving the proactivity and precision of occupational safety management.

Technology Category

• Natural Language Processing: Safety and Robustness
• Machine Learning: Time-Series/Data Streams
• Planning, Routing, and Scheduling: Model-Based Reasoning

Application Category

• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

We propose a generic framework for short-term occupational accident forecasting that leverages safety inspections and models accident occurrences as binary time series. The approach generates daily predictions, which are then aggregated into weekly safety assessments to better inform decision making. To ensure the reliability and operational applicability of the forecasts, we apply a sliding-window cross-validation procedure specifically designed for time series data, combined with an evaluation based on aggregated period-level metrics. Several machine learning algorithms, including logistic regression, tree-based models, and neural networks, are trained and systematically compared within this framework. Unlike the other approaches, the long short-term memory (LSTM) network outperforms the other approaches and detects the upcoming high-risk periods with a balanced accuracy of 0.86, confirming the robustness of our methodology and demonstrating that a binary time series model can anticipate these critical periods based on safety inspections. The proposed methodology converts routine safety inspection data into clear weekly risk scores, detecting the periods when accidents are most likely. Decision-makers can integrate these scores into their planning tools to classify inspection priorities, schedule targeted interventions, and funnel resources to the sites or shifts classified as highest risk, stepping in before incidents occur and getting the greatest return on safety investments.

Problem

• Forecasting short-term occupational accidents using safety inspections
• Evaluating machine learning models for binary time series predictions
• Converting inspection data into weekly risk scores for decision-making

Innovation

• LSTM network for high-risk period detection
• Sliding-window cross-validation for time series
• Binary time series modeling with inspections