Measuring the Opinion of Textile Industries of Punjab (India) towards the Reliability of their Connected Power Systems and its Effects on the Industrial Operations
DOI:
https://doi.org/10.25728/assa.2019.19.1.613Keywords:
Reliability, Power Systems, Questionnaire Schedule, Cronbach’s alpha, Krippendorff’s Alpha, Kruskal Wallis Test, Bonferroni Post Hoc Test, Mann-Whitney U Test, Effect Size, Punjab (India)Abstract
The purpose of this research was to measure the opinion of textile industries of Punjab (India) towards the reliability of their connected power systems of the sole public utility company - Punjab State Power Corporation Limited considering the period from 2010 to 2014. Out of the 400 randomly selected, 148 registered textile industrial units from Central, North and South Punjab, respectively, were responded successfully by conducting the interviews using a questionnaire schedule and dropping questionnaires with the responders, over a period of two years from April 2015-April 2017 which revealed that majority of the industries from the whole Punjab was discontented towards the occurrence of planned outages, unplanned outages and load shedding, work efficiency of staff, condition of infrastructure and tariff structure except the limited number of units from Central Punjab that have shown agreement towards the different aspects of power system reliability. An internal consistency and intra-rater reliability using Cronbach’s alpha and Krippendorff’s Alpha, respectively, calculated, was found very decent. A Kruskal Wallis test was applied and mean rank of Central Punjab was found considerably higher followed by South and North Punjab for the majority of the variables. Dunn’s Bonferroni Post Hoc exposed that a Central Punjab is significantly different from both North and South Punjab but no statistical difference was found between the North and South Punjab. A Kruskal Wallis test statistic (H) and Mann Whitney test statistics (U) was used to calculate the overall effect sizes and the effect sizes between the groups for all the dependent variables, respectively. Out of the forty variables, thirty-three has shown the statistically significant difference from which twenty-eight, three and five were found in the zone of large, intermediate and small effect size, respectively.
