Big data platform for health and safety accident prediction
| Published date | 07 January 2019 |
| Pages | 2-21 |
| DOI | https://doi.org/10.1108/WJSTSD-05-2018-0042 |
| Date | 07 January 2019 |
| Author | Anuoluwapo Ajayi,Lukumon Oyedele,Juan Manuel Davila Delgado,Lukman Akanbi,Muhammad Bilal,Olugbenga Akinade,Oladimeji Olawale |
| Subject Matter | Public policy & environmental management,Environmental technology & innovation |
Big data platform for health and
safety accident prediction
Anuoluwapo Ajayi
Faculty of Business and Law, University of the West of England, Bristol, UK
Lukumon Oyedele
Bristol Enterprise and Innovation Centre, Bristol Business School,
University of the West of England, Bristol, UK
Juan Manuel Davila Delgado
Big Data Analytics Lab, University of the West of England Bristol, Bristol, UK
Lukman Akanbi
Big Data Analytics Lab, University of the West of England Bristol,
Bristol, UK and
Department of Computer Science and Engineering, Faculty of Technology,
Obafemi Awolowo University, Ile-Ife, Nigeria
Muhammad Bilal and Olugbenga Akinade
Big Data Analytics Lab, University of the West of England Bristol,
Bristol, UK, and
Oladimeji Olawale
Faculty of Business and Law, University of the West of England, Bristol, UK
Abstract
Purpose –The purpose of this paper is to highlight the use of the big data technologies for health and safety
risks analytics in the power infrastructure domain with large data sets of health and safety risks, which are
usually sparse and noisy.
Design/methodology/approach –The study focuses on using the big data frameworks for designing a
robust architecture for handling and analysing (exploratory and predictive analytics) accidents in power
infrastructure. The designed architecture is based on a well coherent health risk analytics lifecycle.
A prototype of the architecture interfaced various technology artefacts was implemented in the Java language
to predict the likelihoods of health hazards occurrence. A preliminary evaluation of the proposed architecture
was carried out with a subset of an objective data, obtained from a leading UK power infrastructure company
offering a broad range of power infrastructure services.
Findings –The proposed architecture was able to identify relevant variables and improve preliminary
prediction accuracies and explanatory capacities. It has also enabled conclusions to be drawn regarding the
causes of health risks. The results represent a significant improvement in terms of managing informationon
construction accidents, particularly in power infrastructure domain.
Originality/value –This study carries out a comprehensive literature review to advance the health and
safety risk management in construction. It also highlights the inability of the conventional technologies in
handling unstructured and incomplete data set for real-time analytics processing. The study proposes a
technique in big data technology for finding complex patterns and establishing the statistical cohesion of
hidden patterns for optimal future decision making.
Keywords Big data analytics, Health and safety, Machine learning, Health hazards analytics
Paper type Research paper
1. Introduction
Occupational accidents are things of worry in modern society, especially in construction
sites where a high number of construction activities take place (Zhu et al., 2016). The power
infrastructure delivery sector, for instance, has high incidences of nonfatal occupational
injuries as workers using heavy machinery are confronted with health risks, such as
World Journal of Science,
Technology and Sustainable
Development
Vol. 16 No. 1, 2019
pp. 2-21
© Emerald PublishingLimited
2042-5945
DOI 10.1108/WJSTSD-05-2018-0042
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/2042-5945.htm
2
WJSTSD
16,1
radiation, dust, temperature extremes and chemicals amongst others (McDermott and
Hayes, 2016). According to the UK Health and Safety Executive, a total cost of £4.8bn was
expended in 2014/2015 for workplace injury (HSE, 2016). Similarly, repair costs of buried
communication lines are significant when disrupted during excavations (McDermott and
Hayes, 2016).
Several machine-learning techniques have been used for health and safety risks
prediction in construction. For instance, decision trees (Cheng et al., 2011), the generalised
linear model (Esmaeili et al., 2015) and fuzzy-neural method (Debnath et al., 2016) have all
been used to analyse incident data to reduce accident rates. Techniques, such as the
Bayesian network, were used to quantify occupational accident rates (Papazoglou et al.,
2015), and fuzzy Bayesian networks for damaged equipment analysis (Zhang et al., 2016).
Others are the bow tie representation for occupational risks assessment ( Jacinto and Silva,
2010), and Poisson models for occupational injury impacts modelling (Yorio et al., 2014).
However, a significant problem associated with these existing models is their limited
ability to process large-scale raw data since considerable effort is needed to transform
them into an appropriate internal form to achieve high prediction accuracy (Esmaeili et al.,
2015). Construction accident data are typically large, heterogeneous and dynamic (Fenrick
and Getachew, 2012), nonlinear relationships among accident causation variables
(Gholizadeh and Esmaeili, 2016), imbalance data and appreciable missing values
(Bohle et al., 2015). Besides, these techniques simplify some key factors and pay little
attention to analysing relationships between a safety phenomenon and the safety data
(Landset et al., 2015).
Based on the preceding, the big data technology due to its parallel processing feature
and ability to efficiently handle high dimensional, noisy data with nonlinear relationships,
will be beneficial for health and safety risks analytics in the power infrastructure domain.
Also, the technology will uncover potential factors contributing to accidents in this
domain. The objectives of this study are, therefore, to chart lifecycle stages of occupational
hazards analytics and develop a big data architecture for managing health and
safety risks.
1.1 Big data for health and safety risk analytics
Big data is an emerging technology, which refers to data sets that are many orders of
magnitude larger than the standard files transmitted via the internet (Suthakar et al., 2016).
There is tremendous interest in utilising information in big data for various analytics
(exploratory, descriptive, predictive and prescriptive) to determine future occurrences. Most
importantly, Big data technologies support analytical techniques for occupational health
and safety risk analytics; thus, a system being proposed in this study, named Big Data
Accident Prediction Platform (B-DAPP) offers unparalleled opportunities to minimise
occupational hazards at construction sites. The seamless combination of the following
technologies: big data, health and safety, and machine learning is an outcome of a robust
health and safety risk management tool to help stakeholders in making appropriate
decisions to minimise occupational accidents in power infrastructure projects.
Health and safety risk analytics is dependent on a high-performance computation and
large-scale data storage requiring a large number of diverse data sets of health and safety
risks, and machine-learning knowledge to successfully provide the needed analytical
responsibilities. The data sets, however, are unreliable, unstructured, incomplete and
imbalanced (Chen et al., 2017). Hence, storing the data sets using conventional technologies
and subjecting them to real-time processing for advanced analytics is highly challenging.
A robust technique for finding complex patterns and establishing the statistical cohesion of
hidden patterns in such data sets for optimal future decision making is inevitable. Thus,
motivating the use of big data technologies to address these challenges.
3
Health and
safety accident
prediction
To continue reading
Request your trial