Department

University of Tennessee at Chattanooga. Dept. of Psychology

Publisher

University of Tennessee at Chattanooga

Place of Publication

Chattanooga (Tenn.)

Abstract

Organizations use data to predict future safety incidents through identifying trends and subsequent interventions. However, a lack of data variability can prove fatal to analytics, as it gives no opportunity to capitalize on discrepancies that should exist in everyday working environments. Without the ability to use data to correlate safety incidents and associated factors, targeted safety interventions become more difficult to implement effectively. A solution to this obstacle lies in reassessing current data-collection practices and directly providing an opportunity for greater variance. Thus, we recently worked to enhance the safety recording practices at production facilities of a global threads manufacturing company. Each of the company’s global manufacturing facilities use a five-factor indicator index to maintain record of all plant safety practices—resulting in an overall ‘safety score’ for each facility at the end of every month. However, these records do not sufficiently represent each facility’s practices, as 60% of the indices are measured on a binomial scale. Rather than detailed inventories of safety training and associated details (e.g., the number of employees who participated in each session, assessment results, etc.), each facility indicated whether or not they had distributed assigned safety information in some form. In safety, it is not enough to simply indicate whether or not things are being recorded; it is essential to measure safety elements in order to identify safety incident correlates and implement effective interventions. To develop a more granular system to measure leading safety indicators, we conducted multiple and repeated interviews with SMEs, conducted an audit of all previously required and turned in safety measurement materials, and made note of any consequences associated with failure to report monthly scores from the org’s 32 global manufacturing facilities. The resultant Leading Indicator Index Scale consists of the original 5 factors, each with now 4 levels: “Needs Work,” “Below Standard,” “Standard,” and “Exceeds Standard”, where ratings increase in increments of 5 points (5, 10, 15, and 20, respectively). Data subsequently collected through this new system would be used in correlations to examine the relationships between plants and indices. This data could also be used in conjunction with reported incidents in linear regressions in effort to identify direct predictors of safety incidents, allowing facilities to quickly or even preemptively address safety concerns. In short, the implications of this study on safety analytics are tremendous, as it develops a more acute model organizations can use to audit their safety practices.

Date

10-16-2021

Subject

Industrial and organizational psychology

Document Type

posters

Language

English

Rights

http://rightsstatements.org/vocab/InC/1.0/

License

http://creativecommons.org/licenses/by/4.0/

Share

COinS
 

What do you mean by ‘safe’?: Embedding granularity into a global threads company’s safety recording practices

Organizations use data to predict future safety incidents through identifying trends and subsequent interventions. However, a lack of data variability can prove fatal to analytics, as it gives no opportunity to capitalize on discrepancies that should exist in everyday working environments. Without the ability to use data to correlate safety incidents and associated factors, targeted safety interventions become more difficult to implement effectively. A solution to this obstacle lies in reassessing current data-collection practices and directly providing an opportunity for greater variance. Thus, we recently worked to enhance the safety recording practices at production facilities of a global threads manufacturing company. Each of the company’s global manufacturing facilities use a five-factor indicator index to maintain record of all plant safety practices—resulting in an overall ‘safety score’ for each facility at the end of every month. However, these records do not sufficiently represent each facility’s practices, as 60% of the indices are measured on a binomial scale. Rather than detailed inventories of safety training and associated details (e.g., the number of employees who participated in each session, assessment results, etc.), each facility indicated whether or not they had distributed assigned safety information in some form. In safety, it is not enough to simply indicate whether or not things are being recorded; it is essential to measure safety elements in order to identify safety incident correlates and implement effective interventions. To develop a more granular system to measure leading safety indicators, we conducted multiple and repeated interviews with SMEs, conducted an audit of all previously required and turned in safety measurement materials, and made note of any consequences associated with failure to report monthly scores from the org’s 32 global manufacturing facilities. The resultant Leading Indicator Index Scale consists of the original 5 factors, each with now 4 levels: “Needs Work,” “Below Standard,” “Standard,” and “Exceeds Standard”, where ratings increase in increments of 5 points (5, 10, 15, and 20, respectively). Data subsequently collected through this new system would be used in correlations to examine the relationships between plants and indices. This data could also be used in conjunction with reported incidents in linear regressions in effort to identify direct predictors of safety incidents, allowing facilities to quickly or even preemptively address safety concerns. In short, the implications of this study on safety analytics are tremendous, as it develops a more acute model organizations can use to audit their safety practices.