Journal of Safety Research

Journal of Safety Research

Volume 58, September 2016, Pages 79-87
Journal of Safety Research

Off-road truck-related accidents in U.S. mines

https://doi.org/10.1016/j.jsr.2016.07.002Get rights and content

Highlights

  • Off-road trucks are one of the major sources of equipment-related accidents in the US mining industries.

  • Severe and non-severe injuries are analyzed using a novel clustering-classification method.

  • The accident patterns and trends were identified for all recorded accidents since 2000.

  • The identified accident patterns may play a vital role in the prevention of severe injuries.

Abstract

Introduction

Off-road trucks are one of the major sources of equipment-related accidents in the U.S. mining industries. A systematic analysis of all off-road truck-related accidents, injuries, and illnesses, which are reported and published by the Mine Safety and Health Administration (MSHA), is expected to provide practical insights for identifying the accident patterns and trends in the available raw database. Therefore, appropriate safety management measures can be administered and implemented based on these accident patterns/trends.

Methods

A hybrid clustering-classification methodology using K-means clustering and gene expression programming (GEP) is proposed for the analysis of severe and non-severe off-road truck-related injuries at U.S. mines. Using the GEP sub-model, a small subset of the 36 recorded attributes was found to be correlated to the severity level.

Results

Given the set of specified attributes, the clustering sub-model was able to cluster the accident records into 5 distinct groups. For instance, the first cluster contained accidents related to minerals processing mills and coal preparation plants (91%). More than two-thirds of the victims in this cluster had less than 5 years of job experience. This cluster was associated with the highest percentage of severe injuries (22 severe accidents, 3.4%). Almost 50% of all accidents in this cluster occurred at stone operations. Similarly, the other four clusters were characterized to highlight important patterns that can be used to determine areas of focus for safety initiatives.

Conclusions

The identified clusters of accidents may play a vital role in the prevention of severe injuries in mining. Further research into the cluster attributes and identified patterns will be necessary to determine how these factors can be mitigated to reduce the risk of severe injuries.

Practical application

Analyzing injury data using data mining techniques provides some insight into attributes that are associated with high accuracies for predicting injury severity.

Introduction

Analysis of workplace injuries has been heavily utilized as a means to determine high-risk tasks, prioritize workplace redesign, and determine areas of concern for worker safety in many industries including healthcare, construction, retail and services, and mining (Cato et al., 1989, Drury et al., 2012, Mardis and Pratt, 2003, Moore et al., 2009, Pollard et al., 2014, Schoenfisch et al., 2010, Turin et al., 2001, Wiehagen et al., 2001). While many industries would require injury records from individual companies or insurance providers to perform an analysis, mining is uniquely suited for a more comprehensive injury analysis. An important feature of U.S. mining is the accessibility of injury records. The Mine Safety and Health Administration requires all mine operators and contractors to file a Mine Accident, Injury and Illness Report (MSHA Form 7000-1) for all reportable accidents, injuries, and illnesses incurred at U.S. mining facilities. Reportable illnesses include any illness or disease that may have resulted from work. The database of these reports is available in the public domain and is provided by the National Institute for Occupational Safety and Health (http://www.cdc.gov/niosh/mining/data/default.html). Each entry of the database contains 36 unique attributes including: mine id, mining method, accident date, degree of injury, accident classification, mining equipment, employee's experience and activity, and a narrative briefly explaining the accident. Previous mining research has examined the injury and fatality causes associated with maintenance and repair, haulage vehicles, ingress and egress from mobile equipment, operating underground and surface mining mobile equipment, and other mining tasks (Drury et al., 2012, Moore et al., 2009, Pollard et al., 2014, Reardon et al., 2014, Turin et al., 2001, Wiehagen et al., 2001). Traditional injury data analysis uses counts and cross-tabulations as a means to determine trends in injuries. While this typically yields useful information, more sophisticated data mining techniques may allow for more improved classification of injuries through identification of injury patterns.

Clustering and classification are the two widely used methods of data mining for the purpose of pattern recognition. Clustering is among the unsupervised methods of pattern recognition while the classification is a supervised learning method. By an unsupervised method, one means that the data analyzer does not have any prior hypothesis or pre-specified models for the data, but wants to understand the general characteristics or the structure of the high-dimensional data. A supervised method means that the investigator wants to confirm the validity of a hypothesis/model or a set of assumptions, given the available data (Jain, 2010). Clustering and classification are also called un-labeled and labeled, respectively. In pattern recognition, data analysis is concerned with predictive modeling: given some training data, the prediction task is to find the behavior of the unseen test data. This task is also referred to as learning. Often, a clear distinction is made between learning problems that are (i) supervised (classification) or (ii) unsupervised (clustering), the first involving only labeled data (training patterns with known category labels), while the latter involves only unlabeled data (Duda et al., 2001, Jain, 2010). Clustering and classifications are performed using differing algorithms but may be used together to improve prediction accuracy.

The aim of this research was to gain a better understanding of the factors associating with severe injuries (fatalities and permanent disabilities) in U.S. mining by employing data mining techniques. Clustering and classification were employed for a comprehensive analysis of off-road truck-related accidents and injuries reported to MSHA during a 13-year period (2000–2012). Gene expression programming was used for classification, allowing all injury attributes to be considered and tested to determine which were associated with the highest prediction accuracies. The most explanatory attributes were selected among the available 36 unique attributes in the MSHA database. Then, K-means clustering was used as a means to identify similarity/dissimilarity between the accident records using the selected attributes for the purposes of pattern recognition in the raw data. It should be noted that the goal of this study was not to establish cause–effect relationships between accident attributes and outcomes, but to: (a) use data mining to systematically identify important attributes from MSHA incident reports that are highly associated with the outcomes of accidents (classification), and (b) recognize patterns in the accidents (clustering) given a set of work-related attributes.

Section snippets

MSHA injury data

A dataset comprised of 13 years (2000–2012) of Mine Accident, Injury and Illness Reports was selected beginning with 1/1/2000 (MSHA, 2014). From this dataset, records of severe injuries (fatalities and permanent disabilities) and non-severe injuries associated with off-road trucks were selected. The NIOSH code “minemach-44, all accidents related to off-road mining trucks” was identified to select the records of interest in this study. A total of 5,831 records of injuries (both severe and

MSHA injury data overview

A brief analysis of the MSHA injury data was performed to give an understanding of the types of data available in the injury records. A temporal illustration of off-road truck-related fatalities and disabilities (Fig. 2) shows fluctuations in the fatality and disability counts over the years with a clear decreasing trend. The distribution of accident types is shown in Fig. 3. An analysis of the type of mine operation found that most severe injuries occurred at surface mines that use strips,

Discussion and conclusions

Truck-related fatalities have been previously examined in the literature. An analysis of fatal truck-related accidents during the period 1995–2006 revealed the three most frequent causes of the haul truck-related fatalities as: (i) failure of victims to respect haul truck working area, (ii) failure to provide adequate berms, and (iii) failure of mechanical components (Md-Nor, Kecojevic, Komijenivic, & Groves, 2008). Another study within the period of 1995–2002 (Kecojevic & Radomsky, 2004)

Practical applications

Analyzing the injury data using data mining techniques provides some insight into attributes that are associated with high accuracies for predicting injury severity. Off-road truck-related injuries continue to plague the mining industry resulting in fatalities, permanent disabilities, and other less severe injuries. Many factors contribute to these injuries, and many of these are likely preventable. This analysis revealed that injuries associated with the use of non-powered hand tools for

Saeid R. Dindarloo holds Ph.D., M.Sc. and B.Sc. degrees, all in Mining Engineering, from Missouri University of Science and Technology, USA, and Amirkabir University of Technology (Tehran Polytechnic), Iran. Dr. Dindarloo has 10 years of research and professional experience. He has extensive experience in mine design, planning, and computer application. His current research interests include: mini safety and health, mining machinery, and mining equipment management.

References (27)

  • R. Duda et al.

    Pattern classification

    (2001)
  • C. Ferreira

    Gene expression programming: A new adaptive algorithm for solving problems

    Complex Systems

    (2001)
  • C. Fraley et al.

    Model-based clustering, discriminant analysis, and density estimation

    Journal of the American Statistical Association

    (2002)
  • Cited by (14)

    • Noise exposure and mental workload: Evaluating the role of multiple noise exposure metrics among surface miners in the US Midwest

      2022, Applied Ergonomics
      Citation Excerpt :

      We have developed a conceptual model to demonstrate our hypotheses on how various factors may increase the risk of injury at work (Fig. 1-a). The individual and organizational factors include but not are limited to: age, work experience, work activity, type of mine, and hearing protection device (HPD) use, and have been extensively studied in the mining industry (Dindarloo et al., 2016; Sammarco et al., 2016; Cui et al., 2015; Mitchell et al., 1998). However, there is increasing evidence that occupational noise may be a significant predictor of work-related injuries (Dzhambov and Dimitrova, 2017; Girard et al., 2015; Amjad-Sardrudi et al., 2012; Picard et al., 2008).

    • Evaluation of safety and social indexes in the selection of transportation system alternatives (Truck-Shovel and IPCCs) in open pit mines

      2018, Safety Science
      Citation Excerpt :

      Drury et al. (2012) introduced the haul truck accidents in a pattern via data analysis in order to predict the accident model. Dindarloo et al. (2016) evaluated the severity of injuries caused by mine haul trucks through hybrid clustering classification. They made a cluster definition for different types of accidents, including location, the experience of the victim and severity.

    View all citing articles on Scopus

    Saeid R. Dindarloo holds Ph.D., M.Sc. and B.Sc. degrees, all in Mining Engineering, from Missouri University of Science and Technology, USA, and Amirkabir University of Technology (Tehran Polytechnic), Iran. Dr. Dindarloo has 10 years of research and professional experience. He has extensive experience in mine design, planning, and computer application. His current research interests include: mini safety and health, mining machinery, and mining equipment management.

    Jonisha P. Pollard M.S., CPE is a Research Engineer at the National Institute for Occupational Safety and Health in the Office of Mine Safety and Health Research. She joined NIOSH while pursuing her Master's degree in Bioengineering from the University of Pittsburgh in 2007. During her time at NIOSH, she has been involved in numerous research studies and has published research on knee injuries in mining, the effect of cap lamp lighting on balance, the effect of kneepads on balance, injuries due to maintenance and repair work in mining, locomotion in restricted workspaces and a variety of other topics. Ms. Pollard enjoys conducting field and laboratory research to positively impact the health and safety of mine workers.

    Elnaz Siami Irdemoosa obtained her B.Sc. in Mining Engineering and her M. Sc. in Rock Mechanics from Amirkabir University of Technology (Tehran Polytechnic). She is currently a Ph. D. candidate at Missouri University of Science and Technology in the field of Geological Engineering. Her research interests include underground design and construction, tunneling, construction management, and geophysical method.

    View full text