Cross-sectional analysis of incident causations within the South African civil construction industry

Allsopp, Shane (2020-03)

Thesis (MEng)--Stellenbosch University, 2020.


ENGLISH ABSTRACT: The civil construction industry in South Africa contributes greatly to the occurrence of work-related incidents in the country, often leading to injury or fatality. In order for South Africa to reduce the high incident rates in this industry, it must first become aware what is causing these incidents. The study sets out to provide information regarding the causation of incidents within the South African civil construction industry, identifying leading incident causation factors in terms of human errors, workplace factors and organisational factors. Relationships between these incident causation factors develops understanding of the failure pathway which leads to an incident occurring. Understanding incident causations can be done through the analysis of incident reports. The incident reports are gathered from three South African civil construction companies who shall remain anonymous. Each individual incident report is analysed using an incident causation framework (ICF) adapted from Jude Bonsu’s work on the South African mining industry (Bonsu et al., 2016). A cross-sectional analysis is then done across all incident reports, whereby leading incident causation factors and relationships between incident causation factors are identified. Prior to performing analysing the incident reports, motivation is given of the need for this study in the South African civil construction industry. The motivation is given through the comparison of incidents between this industry and the South African mining industry. The results of this comparison found that the South African mining industry has successfully reduced both injury and fatality frequency rates, whilst the civil construction data had no statistical significance. Lack of statistical significance indicated a lack of control over incidents. The South African civil construction has managed to reduce injury frequency rates; however, they were experiencing rates far higher than those of the mining industry. The findings of this study presented that common leading incident causations were occurring for the three companies. For all three companies; mistakes are found to be the leading human error; controlled work environment (CWE) the leading workplace factor; and hazard identification the leading organisational factor. Relationships were also established that linked the workplace factors to the human errors that they cause and organisational factors to the workplace factors they cause. The relationships were analysed, and it was found that reasoning could be given to the relationships occurring between each of the incident causation factors. The research concluded that it had both achieved the objectives set out as well as provided benefits that can be applied to the South African civil construction industry. Recommendations were also made to promote further studies and application of this study to the South African civil construction industry.

AFRIKAANSE OPSOMMING: Die siviele konstruksiebedryf in Suid-Afrika dra grootliks by tot werkverwante voorvalle in die land, wat dikwels lei tot beserings of sterftes. Om Suid-Afrika se hoë aantal voorvalle in hierdie bedryf te verlaag moet mense eers bewus word van wat hierdie voorvalle veroorsaak. Hierdie studie beoog om inligting te verskaf rakende die oorsake van voorvalle in die Suid-Afrikaanse siviele konstruksiebedryf, deur die vernaamste oorsaakfaktore in terme van menslike foute, werksplekfaktore en organisatoriese faktore te identifiseer. Die verwantskappe tussen hierdie oorsaakfaktore ontwikkel begrip van die oorsake wat lei tot ‘n voorval. Om die oorsake van voorvalle te verstaan, moet verskillende voorvalverslae ontleed word. Voorvalverslae van drie Suid-Afrikaanse siviele konstruksiemaatskappye wat anoniem sal bly is ontleed. Elke voorvalverslag is geanaliseer en ontleed met behulp van ‘n insident oorsaak raamwerk wat aangepas is uit Jude Bonsu se werk binne die Suid-Afrikaanse mynbedryf. Daarna word ‘n dwarsnit-analise oor alle voorvalverslae gedoen, waaruit leidende oorsaakfaktore en verwantskappe tussen voorvaloorsaakfaktore geïdentifiseer word. Voordat die voorvalverslae ontleed word, word ‘n motivering gegee vir die behoefte aan hierdie studie binne die Suid-Afrikaanse siviele konstruksiebedryf. Die motivering word gegee deur die voorvalle tussen hierdie bedryf en die Suid-Afrikaanse mynbedryf te vergelyk. Uit die resultate is bevind dat die Suid-Afrikaanse mynbedryf die beserings- en die sterftefrekwensiesyfers suksesvol verlaag het, terwyl die gegewens vir die siviele konstruksiebedryf geen statistiese betekenis gehad het nie. ‘n Gebrek aan statistiese betekenis dui op ‘n gebrek aan beheer oor voorvalle. Die Suid-Afrikaanse siviele konstruksiebedryf het dus daarin geslaag om die frekwensie van beserings te verminder, alhoewel hul egter koerse ervaar wat veel hoër is as die mynbedryf. Die bevindinge van hierdie studie het aangedui dat die mees algemene oorsake van voorvalle by al drie maatskappye voorkom. Vir al drie maatskappye is gevind dat; foute die grootste menslike oorsaak is; beheerde werksomgewing die grootste werkplekfaktor is; en gevaaridentifisering die leidende organisatoriese faktor is. Verwantskappe is ook bewerkstellig wat die werkplekfaktore koppel aan die menslike oorsake wat dit veroorsaak en organisatoriese faktore met die werkplekfaktore wat dit veroorsaak. Die verwantskappe is geanaliseer en daar is gevind dat daar ‘n redenasie is vir die verwantskappe tussen elk van die oorsaakfaktore. Die gevolgtrekking uit die navorsing is dat dit beide die uiteengesette doelstellings bereik het, sowel as voordele bied wat binne die Suid-Afrikaanse siviele konstruksiebedryf toegepas kan word. Aanbevelings is ook gemaak vir verdere studies en om die toepassing van hierdie studie binne die Suid-Afrikaanse siviele konstruksiebedryf te bevorder.

Please refer to this item in SUNScholar by using the following persistent URL:
This item appears in the following collections: