Factors Influencing the Adoption of Smart Predictive Maintenance in the Cement Industry in Kenya

Abstract

The purpose of this research study was to identify the factors influencing the adoption of smart predictive maintenance in the cement industry in Kenya. The study was guided by the following research questions: How does relative advantage influence the adoption of smart predictive maintenance in the cement industry in Kenya? How does compatibility influence the adoption of smart predictive maintenance in the cement industry in Kenya? How does complexity influence the adoption of smart predictive maintenance in the cement industry in Kenya? This research adopted a descriptive research design to answer the study’s objective. The target population is composed of 108 respondents, holding management, engineering, and supervisorial roles, distributed in the cement manufacturing companies located in Kenya. The population sample size that was adopted for this study was 85 respondents based on the target population. A questionnaire was used for data collection. The resultant data was analysed using descriptive statistics such as mean and standard deviation while inferential statistics involved the determination of relationships and variances among the variables. For this study, multinomial regression analysis was used. The data analysis tool used was Statistical Package for the Social Sciences (SPSS) and thereafter, the data was presented in graphs, tables, and figures. The study found that among the constructs measuring relative advantage as a factor influencing the adoption of predictive maintenance in the cement industry, the respondents were generally in strong agreement with the measures. The relative advantage that the study determined to have an impact included enhanced productivity, better warehouse management, improved competitive advantage, job effectiveness, just-in-time maintenance, cost reduction, and equipment uptime. Further, the deduction from the study findings was that many cement firms look for technology that promises a competitive advantage that not only increases profitability but also lowers operational costs and improves efficiency. Further relative advantage was determined, independently, not to be a statistically significant predictor of the extent of adoption of Smart Predictive Maintenance technologies [ᵡ2(2) =5.075, p>0.05]. This study considered compatibility as an important anchor in the adoption of smart predictive maintenance technology by the cement industry in Kenya. The findings of the study suggest that for most cement firms, the likelihood of adopting a smart predictive maintenance technology was dependent on the extent to which it will require them to adjust their existing operational and organizational routines. As such, technologies with less requirement for adjustment were easily adopted and the probability of adoption of technology was significantly reduced if it was deemed to be incompatible with existing technological infrastructure. Additionally, the type of the technology, its compatibility with existing processes, security features it confers, user-friendliness of the technology as well as its quality and cost also determined its ease of adoption. The study also established that compatibility, independently, was a statistically significant predictor of the extent of adoption of Smart Predictive Maintenance technologies [ᵡ2(2) =67.705, p<0.05]. The complexity of the system with respect to the installation process, extraction of information from the system, maintenance, and preparation models for maintenance activities was determined to be an important factor influencing decisions of whether to adopt a new technology or not. The consensus among study participants was that the ease of adoption of smart predictive maintenance technology had an inverse relationship with the degree of complexity of smart predictive maintenance technology. Further, the odds of adoption of the technology also increased it easily communicates the direct and indirect benefits it promises such as reducing unplanned downtime, optimizing the use of resources, and accelerating the speed of diagnosis of potential risks to the processes of the firm. Further, complexity, independently, was found to be a statistically significant predictor of the extent of adoption of Smart Predictive Maintenance technologies [ᵡ2(2) =37.290, p<0.05]. The research study concluded that while smart predictive maintenance technology help firms move away from reactive and/or preventive approach to keeping the critical machinery up and running, their adoption is still hinged on the relative advantage they promise, their degree of ease of use, and the extent to which the technology is perceived to be consistent with previous experiential, technical, and operational dynamics of the firms. The study recommends that the firms need to invest in technology that accounts for the real-time flow of data which accords them the capability to predict repair needs prior to their occurrence and allows the firm to establish a maintenance plan thereby avoiding downtime and saving resources by minimizing costs of maintenance.