Impact of Implementation of Behaviour-Based Safety Programme on At-Risk Behaviours in the Nigerian Chemical Process Industry

Impact of Implementation of Behaviour-Based Safety Programme on At-Risk Behaviours in the Nigerian Chemical Process Industry

¹Esang Lazarus Esitikot, ²Mary Ubong Umoh, ³Akaninyene Edet Ekong, ⁴Utibe Amos Ofon, ⁵Anthony Akadi, ⁶Clement O. Obadimu

^1,3Highstone Global University, Texas, USA

^2,4,5,6Institute of Health, Safety, Security and Environment Studies, University of Uyo, Nigeria

DOI: https://doi.org/10.51584/IJRIAS.2024.911014

Received: 26 October 2024; Accepted: 12 November 2024; Published: 04 December 2024

ABSTRACT

Over the years, several strategies have been deplored to improve safety performance in organizations and eliminate accidents. Part of the strategies involved improved engineering designs, personnel training and deployment of safety management systems. While these strategies resulted in significant reduction in accidents rates in the different industries, they have not resulted in complete elimination of accidents. Behaviour-based safety (BBS) programme was introduced to address accidents from human frailties through a systematic process of improving safety behaviours. The process uses behavioural psychology and employee involvement to prevent accidents believed to be caused mainly by human errors. Despite the resources expended in implementation of BBS, researchers do not agree on BBS as an effective instrument for improvement of health and safety management in organizations. This research was undertaken to determine how BBS implementation affects at-risk behaviour. Specifically, the study aimed to establish whether implementation of BBS programme led to reduction in at-risk behaviours among workers in the chemical process industry. The research was conducted using questionnaires distributed to 50 workers (of different job levels and categories) in the Nigerian chemical process industry. The workers were selected through a stratified random sampling technique. The questionnaires evaluated key elements of BBS as implemented by different organizations in the industry. The data obtained were analyzed using descriptive statistics and correlation. The study established that deployment of BBS was effective in reducing at-risk behaviours. The implication is that for organizations that are having challenges with unsafe acts, at-risk behaviours, and accidents due to human errors and mistakes, BBS can be an effective tool to deploy to transform the culture. Though this research established that BBS contributes to reduction in at-risk behaviours, further study is recommended to address the proportion of at-risk behaviours that are reduced by the implementation of BBS, and to identify the impact of BBS programme focused on organisation’s management. This would enable organizations implementing BBS programme to proactively project the expected gains from programme deployment.

Keywords: At-risk behaviour, behaviour-based safety, health, safety, safety management, loss prevention observation.

INTRODUCTION

Over the years, organizations in the chemical process industry have applied several approaches to drive achievement of its vision of “zero hurt”. The “zero hurt” philosophy was based on the perception that all accidents are preventable and so having a workplace where no worker gets hurt should be the only acceptable safety goal [1, 2]. The background is that accidents are considered as the products of at-risk behaviours and unsafe conditions [5]. The approaches applied included improved engineering designs, incident reporting and investigation, plant inspections, personnel training, proactive reporting, and implementation of health, safety and environment management system [1, 4]. Each approach matured over time resulting in year-on-year improvements in safety performance. With time, it was observed that the accident rate levelled out with no further significant improvement in safety performance due to human components in accident causation [2, 3, 5, 6]. This necessitated the introduction of behaviour-based safety as a tool to positively influence human behaviours and eliminate at-risk behaviours.
Behaviour-based safety utilizes behavioural science to drive improvements in the safety culture of organisations [3]. The outcome of such improvement, as [3] noted, include improved safety performance and increased worker productivity. Implementation of behaviour-based safety involves strategies such as awareness of the programme among the workforce, assessment of prevailing safety culture, observation of worker behaviour with reference to social, economic, cultural, and legal requirements, and interaction with the worker involved to share observation feedback, solicit his or her perspectives on observations made, reinforce positive behaviours and align on ways of improving task execution in future [3, 5, 7].

A key component of behaviour-based safety is loss prevention observation (LPO), a typical strategy used by organizations to observe their workers while performing tasks so as to identify areas requiring improvements as part of the effort to drive behavioural change. LPO typically involves having an observer (a person not involved with a task) watch an observee (a worker performing a task); and comparing the observee’s behaviours (such as use of personal protective equipment, lifting techniques, reference to task procedure at worksite) with organization’s accepted safety standards or practices [3, 7, 8, 9]. The focus is on establishing whether the observee complied with the applicable organization’s recommended standards, guidelines or requirements for safe performance of the task [7, 9, 10]. A post-assessment review and feedback is conducted between the observee, observee’s supervisor, and observer to reinforce observed positive safety behaviours; agree on reasons for non-adherence to the standards or practices, if deviation to standards was observed; develop action plans to address any observed non-adherence to standards to prevent similar at-risk behaviours in future. Finally, key learnings from the observations are shared with other workers for continuous improvements.

Though different organizations have programmes for assessment of the effectiveness of LPO implementation, the assessment focuses on workforce involvement in the programme and documentation reviews and does not assess impact of the programme on the reduction in at-risk behaviours as part of the effort to improve organizational safety culture, or how the presence of an observer may adversely or positively impact the performance of a task [8], hence the need for this research.

Theoretical/Conceptual Framework

The theoretical/conceptual underpinning of this study is based on behaviourism theory. The theory, postulated by Watson in 1913, has it that there is minimal disruption between the worker and organizational environmental characteristics and that all behaviours are acquired through conditioned interaction with the environment [11]. The goal is that through repetition and training, positive reinforcement and punishment, together with the individual’s motivational state and controlling stimuli, the desired behaviour can become a habit. Through observation, the desired behaviour gets positive reinforcement while the undesired behaviour is awarded negative reinforcement [3, 10, 12].

Behaviour-based safety, therefore, is drawn from the concepts of using colleagues, peers and worker supervisor to repeatedly expose the worker to safe behaviours stimuli. Such exposure acts as a strong group influence to achieve interdependent safety culture that involves every member of the organization [3, 13].

As [3] noted, the basics of accident causation philosophy is that most of the industrial accidents are due to unsafe acts or unsafe conditions or a combination of both. A deep dive at the root causes of accidents shows that human behaviour is the primary cause of undesirable working conditions that ultimately leads to accidents [10]. According to [3], “Numerous studies in this area have revealed that most of these accidents are attributed to casual and careless at-risk behavior of employees in the organisation. Industrial organisations with well-established organisational culture, despite having installed integrated safety management systems and the best compliance practices to remove all accidents, have realized that some incidents and accidents do appear to stubbornly resist all these efforts to remove them.” The implication is that by influencing human behaviours, at-risk behaviours can be reduced or eliminated [3, 10, 13].

METHODOLOGY

Study Design: A multiple case study design was adopted for this study as [14 – 16] identified it as suitable for answering a research question involving multiple cases. The design is also suitable when appropriate for evaluating the similarities and contrast between two or more cases [14 – 16]. Since this study required collecting data from several locations and organizations in different contexts, the multiple case study approach was fit for the study.

Study Population: Population is the set of individuals of interest to the researcher [17]. The study population was workers in the chemical process industry in Nigeria. The industry was considered appropriate for the study as behaviour-based safety has been implemented as an instrument for improving safety culture through reduction of at-risk behaviour [18].

Selection Criteria: Sample is a set of individuals selected to represent the population in a study [17]. A sample can be selected from an accessible population which is a portion of the target population [19]. The locations and organizations selected for the study were and oil and gas production, processing and/or export facilities that run 14-day hitch rotation; hence the workers on duty were the accessible population. Other workers were off duty and could not be readily accessed. 50 workers were selected for the study. Though larger sample size could have improved reliability [20], an advantage of quantitative method is in the use of smaller groups of respondents to make inferences about larger groups [21].

Sampling Technique: To achieve rich data for the study, a stratified random sampling technique was adopted to select respondents to increase sample representativeness [22]. To ensure a variability of 10% at 5% margin error based on Watson’s Table for Finding a Base Sample Size [23], the researcher followed up with respondents to ensure the return of all questionnaires shared. This maximized data gathered.

Research Instrument: Data collection was done using questionnaire. As [24] recommended, the questions were phrased in such a way that the respondent understood the objective of each question. Prior to deployment of the questionnaires for the study, the clarity of the questions was tested through a pilot test involving friends and colleagues and concerns raised in the form of content, grammar, ambiguity, and format were addressed. The questionnaire was designed to obtain a fair representation of the perception of all categories of workers [18, 24] using a five-point Likert-type scale (Strongly disagree = 0; Disagree = 1; Neutral = 2; Agree = 3; Strongly Agree = 4). The number of questions was limited to 20 as questionnaire length is inversely proportional to response rate [25] while a 5-point scale (not less or more) was chosen to align with [26] recommendation on the need to strike a balance between meeting data collection objective, increasing the response rate and accuracy, and refining the data collected. The questions were phrased such that “Strongly disagree” indicated negative relationship between applicable variable whereas “Strongly agree” indicated a positive relationship. The ‘Neutral’ option was given for those without sufficient information to justify an opinion, were indifferent to the subject, or thought the good and bad points were about equal [26]. The questionnaire design limited respondents to just the research subject and comprised of concise closed-ended questions for easy statistical analysis [20, 24] and limitation in expected response details [25]. As [25] recommended, the questions were in logical groups and contained the relevant information needed to meet the research objective.

Procedure: The way research participants would be treated and how data obtained from the research would be managed is important [27] and so was considered in this study. The issues considered included voluntary participation, well-informed consent, and data confidentiality. These were communicated to the respondents through a cover letter and participant’s consent form included in each pack of questionnaire.

Data Analysis: One of the steps in a research procedure is consideration of how the variables measured would be manipulated and regulated to achieve the aim of the study [28]. Descriptive statistics and correlation were identified as suitable for this study [19] and were used to analyze data obtained for the study. These methods consist of graphical and numerical techniques for summarizing data for easy understanding [29].

Ethical Approval: The right of the study participants to withdraw from the study at any time, without the need to provide any justification, was part of the informed consent process. No form of inducement was used to recruit participants for the study. The names of the participants or organizations the work for were not included on the questionnaire as part of effort to maintain the confidentiality of research participants.

RESULTS

Table 1 shows a summary of the demographic characteristics of all the respondents in the research.

Table 1: Summary of Demographic Characteristics of Respondents

Figure 1 compares supervisors’ responses with the responses of other (technical) workers. This helps to show whether there is any difference between supervisors’ level of awareness, training and involvement in BBS observations compared to other workers.

Figure 1: Comparison of Supervisor and Technical Workers Awareness and Involvement in BBS Observation

Table 2 compared the responses from supervisors (on the effectiveness of behaviour-based safety [BBS] observation in relation to reduction in at-risk behaviour) with those from other workers. This was to help determine whether the perception of BBS observation effectiveness was dependent on the job status or level of the respondents.

Table 2: Comparison of Supervisors and Technical Workers Perception of BBS Effectiveness

DISCUSSION OF FINDINGS

To confirm that the data obtained in this study were not due to sampling error, the Excel CHITEST function was used to test the difference in the typical data for supervisors and other (technical) workers. Thus, the probability of the differences between the observed (actual) frequencies and expected frequencies were calculated. Table 3 shows the test of independence for the responses on how LPO contributed to reduction in at-risk behaviour. Generally, in CHITEST, a probability (p) of 0.05 or less is considered to be significant. Thus, with a probability (p) of 0.0023 (which is far less than 0.05), it was clear that the difference was not due to sampling error.

Table 3: Test of Independence of Responses to Reduction in At-Risk Behaviour.

The following factors were considered while testing for the relationship between implementation of BBS and reduction in at-risk behaviours:

Fo – Observed frequency
Fe – Expected frequency
df – Degree of freedom
r – The number of levels of the first independent variables
c – The number of levels of the second independent variables
α = 0.05
χ²– Chi- square
χt² – Critical value (from Chi-square statistical table)

Null Hypothesis, 1H0: Implementation of BBS does not result in reduction in at-risk behaviour

Hypothesis of relationship, 1H1: Implementation of BBS results in reduction in at-risk behaviour.

The data in table 4 were used for the test.

Degree of freedom, df = (r – 1)(c-1), where r =2 and c =5

df = (2-1)(5-1)=4.

χt² = X²_0.05 = 9.49

Table 4: Computation of Chi-square Critical Computed Value (X²c) from Table 3.

Since X²_o is greater than χt² (that is, 14.53 > 9.49), the null hypothesis (1H0) was rejected while the alternative hypothesis (1H1) that implementation of BBS results in reduction in at-risk behaviour was accepted. This was in line with the research findings of [30] who observed an increase in safe behaviours of workers after the consequent intervention and implementation of BBS in the activities of eighteen manufacturing organizations in the United States of America. This finding also agreed with the conclusion of many previous studies [18, 31 – 33] that BBS is effective as a means of reducing at-risk behaviours in organizations. The finding was also consistent with [34, 35] findings on the positive effects of feedback in motivating safe behaviours.
In the experiment conducted by [36], it was concluded that participants that conducted safety observations on other participants were more safety-conscious than participants that did not participate in the safety observation process. 88% of the respondents in this research either agreed or strongly agreed to have served as observer while 92% of the respondent agreed or strongly agreed to have served as observee (figure 1). In line with [36] findings, it could be concluded that the high percentage of perception of reduction in at-risk behaviour was due to the high level of involvement of the workforce in the BBS programme [18]. The implication is that to rely on BBS as an instrument for safety improvement, organizations need to implement programmes that enhance increased involvement of workers in BBS observations.

It is, however, unclear from this study, the proportion of at-risk behaviour that is addressed by the implementation of BBS programme.

CONCLUSION

Based on previous studies, there is no consensus among researchers on the effectiveness of BBS in improving safety management [37, 38]. While some researchers claim that BBS implementation contributes significantly to reduction in at-risk behaviours [4, 7, 9, 31, 39, 40], others see the programme as resulting in only temporary behavioural changes, hence adding no real value to the long-term effectiveness of safety and health management [41]. The controversy revolves mainly around the conclusions of [42] and [43] that human errors are the source of most accidents, and [44] and [45] recommendations on use of people-focused approach to eliminate human errors.

This research, however, established that implementation of BBS resulted in the reduction of at-risk behaviours in line with the research conclusions of [4, 7, 9, 18, 39, 44, 46]. This shows that BBS programme positively affected health and safety management in the chemical process industries, similar to the conclusions of [4, 7, 9, 18, 42]. The implication is that to change the safety culture of an organization, a change in behaviour is necessary since safe behaviour is a critical part of effective health and safety management [9, 42, 47]. As [31] noted, it is behaviour that turns systems and procedures into reality, while culture in turn drives safe behaviour, the focus of BBS [37]. Hence, organizational leaders should adopt BBS as a strategy for eliminating at-risk behaviour to improve organizational safety culture.

RECOMMENDATIONS

Recognizing that the implementation of BBS helps in reduction of at-risk behaviours to improve organizational safety culture, there is the need for increased awareness among the workforce on the essence of the programme. Such awareness can be done using existing safety meetings or dedicated training programmes. To show organizational leadership commitment, the awareness should be led by the most senior management representative at each site. The awareness campaign can be an opportunity to reassure the workforce that deficiencies identified during behavioural observations would not constitute avenues for punitive actions. Such reassurance may enhance increased participation in the programme to maximize the gains.

The scope and focus of BBS should be reviewed and implementation guide revised to distinguish between willful at-risk behaviours versus behaviours that stem from underlying systemic issues in each organisation. This data should be tracked as it can provide a guide where safety efforts and resources should be targeted. The tracking should include monitoring and addressing defective tools and equipment identified during behavioural observations to showcase organizational leadership commitment to using BBS as a tool for safety improvement.

Finally, though this research has established that BBS contributed to the reduction in at-risk behaviours, further study is required to address the proportion of at-risk behaviours that can be reduced by the implementation of BBS, and to identify the impact of BBS programme focused on organisation’s management. This would provide the data required for new organisations planning to implement BBS programme to project the expected gains.

REFERENCES

1. Wang. Y. & Griffis, F. (2018). The theory of zero incident safety management. Journal of Civil, Construction and Environmental Engineering, 3(3), 83-98. https://doi.org/10.11648/j.jccee.20180303.15
2. Fabiano, B. (2017). Loss prevention and safety promotion in the process industries: issues and challenges, Process Safety and Environmental Protection, 110, 1-4. https://doi.org/10.1016/j.psep.2017.08.030
3. Kaila, H. L. (2018). Behavioural intervention in safety management of Indian Corporates. Journal of Psychosocial Research, 13(1), 165–188. https://doi.org/10.32381/jpr.2018.13.01.17
4. Hussain, U., Shoukat, M. H., & Shamail Haider, M. (2019). Analysis of Safety Awareness, Accident Prevention and Implementation of Behavior Based Safety Program in Energy Utility. 2019 6th International Conference on Frontiers of Industrial Engineering (ICFIE), Frontiers of Industrial Engineering (ICFIE), 2019 6th International Conference On, 84–88. https://doi.org/10.1109/ICFIE.2019.8907683
5. Reason, J., 1997. Managing the risks of organisational accidents. Aldershot: Ashgate
6. HSE, 1997. Successful health and safety management. HSE: London
7. Yang, E., Kim, Y., & Rodgers, C. (2024). Effects of a behavior-based safety observation program: Promoting safe behaviors and safety climate at work. Work, 77(1), 133–145. https://doi.org/10.3233/WOR-220465
8. Zhangreview, T. (2024). The impact of observers on peoples behaviour. Lecture Notes in Education Psychology and Public Media 50(1), 243-248. https://doi.org/10.54254/2753-7048/50/20240957
9. Kaila, H. L. (2014). Emerging issues and outcomes of behavior based safety (BBS) implementation. Journal of the Insurance Institute of India, 2(1), 98–102
10. Baharuddin, A., Fachrin, S. A. & Putri, W. E. (2023). Behavior based safety implementation using the DO IT method at Pertamina in Makassar City. Diversity, 30–36. https://doi.org/10.24252/diversity.v4i1.40918
11. Kwon, H. R., & Silva, E. A. (2020). Mapping the landscape of behavioral theories: systematic literature review. Journal of Planning Literature, 35(2), 161-179. https://doi.org/10.1177/0885412219881135
12. Kadau, I. (2022). Behaviourism theory in teaching and learning English as a second language in primary schools. International Journal of English and Literature, 13(3), 33-38. https://doi.org/10.5897/IJEL2022.1511
13. Spigener, J., Lyon, G., & McSween, T. (2022). Behavior-based safety 2022: today’s evidence. Journal of Organizational Behavior Management, 42(4), 336-359–359. https://doi.org/10.1080/01608061.2022.2048943
14. Lederer, M., Kurz, M., & Lazarov, P. (2017). Usage and suitability of methods for strategic business process initiatives: A multicase study research. International Journal of Productivity Management and Assessment Technologies, 5(1), 40–51. https://doi.org/10.4018/IJPMAT.2017010103
15. Saunders, M. N. K., & Rojon, C. (2015). There’s no madness in my method: Explaining how your research findings are built on firm foundations. Coaching: An International Journal of Theory, Research and Practice, 7, 74–83. https://doi.org/10.1080/17521882.2014.889185
16. Yin, R. K. (2014). Case study research: Design and methods (5th ed.). Sage.
17. Mcqueen, R. A. & Knussen, C., (2006). Introduction to research methods and statistics in Psychology. Pearson Education Limited: Essex.
18. Agwu, M. O., 2013. Perception survey of the impact of behaviour based safety on accident in the Bonny NLNG construction project, Nigeria. British Journal of Management, Economic and Trade, 3(1), 48-59. https://doi.org/10 .9734/BJEMT/2013/2391
19. Vander stoep, S. W. & Johnston, D. D. (2009) Research methods for everyday life blending qualitative and quantitative approaches. Jossey-Bass: San Francisco
20. Dawson, C., 2002. Practical research methods. UBS Publishers’ Distributors: New Delhi
21. Holton, E. H., & Burnett, M. B., 1997. Qualitative Research Methods. In R. A. Swanson, & E. F. Holton (Eds.), Human resource development research handbook: Linking research and practice. Berrett-Koehler Publishers: San Francisco.
22. Howell, C.R. et al. (2020). Area based stratified random sampling using geospatial technology in a community-based survey. BMC Public Health 20, 1678 (2020). https://doi.org/10.1186/s12889-020-09793-0
23. Watson, J., 2001. How to determine a sample size: tipsheet #60. Penn State Cooperative Extension: Pennsylvania.
24. Jenn N. C. (2006). Designing a questionnaire. Malaysian Family Physician, 1(1):32-5
25. Eiselen, R. J. et al. (2005. Analysis survey data using SPSS13: A workbook. University of Johannesburg: Johannesburg.
26. Currie, D., 2005. Developing and applying study skills. CIPD: London
27. Cohen, L. et al. (2004). Research methods in education, 5th edition. Routledge-Falmer: London
28. Gravetter, F. J. & Forzano, L. B., 2009. Research methods for the behavioural sciences, 3rd edition. Wadsworth Cengage Learning: Belmont
29. Burns, R. B., 2000. Introduction to research methods, 4th edition. Sage Publications: London.
30. Grindle, A.C. et al. (2000). Behavioral Safety Research in Manufacturing Settings: A Review of the Literature. Journal of Organizational Behavior Management, 20(1), pp 29 – 68.
31. Hopkins, A. (2006). What are we to make of safe behaviour programs? Safety Science, 44, 583-597. https://doi.org/10.1016/j.ssci.2006.01.001
32. Lyon, B.K. (2001). Behavioral sampling of theme park operators. Professional Safety, 46(10), 35-42
33. Veazie, R. (1999). Foundation principles: keys to success of a behavior-based safety initiative. Professional Safety, 44(4), 24-26
34. Komaki, J. et al. (1978). A behavioral approach to occupational safety: pinpointing and reinforcing safe performance in a food manufacturing plant. Journal of Applied Psychology, 63, 424-445.
35. Eckenfelder, D. J. (2004). Behavior Based safety: a model poisoned by the past; based on obsolete thinking, behavior based safety. Risk & Insurance, 15 (12), 65.
36. Sasson, J. R., & Austin, J. (2004). The effects of training, feedback, and participant involvement in behavioral safety observations on office ergonomic behavior. Journal of Organizational Behavior Management, 24(4), 1–30. https://doi.org/10.1300/J075v24n04_01
37. Flemming, M. & Lardner, R., 2002. Strategies to promote safe behaviour as part of health and safety management systems, contract research report 430/2002 for the UK Health and Safety Executive. HSE: London
38. Li, Z., Zhu, Y., Zhao, L., & Zhao, Y. (2024). Safety decision of autonomous driving behavior based on Bayesian network and its verification method. 2024 IEEE 24th International Conference on Software Quality, Reliability, and Security Companion (QRS-C), Software Quality, Reliability, and Security Companion (QRS-C), 2024 IEEE 24th International Conference on, QRS-C, 145–153. https://doi.org/10.1109/QRS-C63300.2024.00029
39. Krause, T. R. et al. (1999). Long-term Evaluation of a bhaviour-based method for improving safety performance: a meta-analysis of 73 interrupted time series replications. Safety Science, 32(2), 1-18
40. Ayuni, M. Q., Yusuf, M. & Dwiyanti, E. (2022). Performance analysis of the behavior based safety program in reducing occupational accident rates. The Indonesian Journal of Occupational Safety and Health, 11(2), 275–284. https://doi.org/10.20473/ijosh.v11i2.2022.275-284
41. Agraz-Boeneker, R. & Haight, J.. (2007). An examination of observations and incidence rates for a behavior based safety program. Journal of SH&E Research. 4. 1-22. https://doi.org/10.3320/1.2758702.
42. Henrich, H. W., 1941. Industrial Accident. Preventian: A Scientific Approach, 2nd edition. McGraw-Hill: London
43. Wagenaar, W. & Groeneweg, J. (1987). Accidents at sea: multiple causes and impossible consequences. International Journal of Man-Machine Studies. 27. 587-598. https://doi.org/10.1016/S0020-7373(87)80017-2.
44. Cooper, M.D. (2003). Behavior based safety still a viable strategy. Safety & Health, April 2003, 46–48.
45. Geller, E. Scott. (2005). Behavior-Based Safety and Occupational Risk Management. Behavior modification. 29. 539-61. https://doi.org/10.1177/0145445504273287.
46. Hojo, R., Sujino, T., Bordlein, C., Koremura, Y., & Shimizu, S. (2021). Risk distribution method from worker behavior measurement using Behavior-Based Safety procedure. 2021 IEEE 3rd Global Conference on Life Sciences and Technologies (LifeTech), Life Sciences and Technologies (Life Tech), 2021 IEEE 3rd Global Conference On, 35–36. https://doi.org/10.1109/LifeTech52111.2021.9391969
47. Jasiulewicz-Kaczmarek, M. et al. (2015). Behaviour based intervention for occupational safety – case study. Procedia Manufacturing, 3, 4876-4883. https://doi.org/10.1016/j.promfg.2015.07.615.