A case study: utilizing Cross-industry standard process for data mining to monitor production costs

Abstract

Manufacturing companies strive for perfection to strengthen competitive edge in today’s global markets. Manufacturers seek to maximize profit and minimize waste by adopting real-time assessment tools to monitor production flow, labor, and impact of improvements. The purpose of the research is to introduce data analytic process and framework in order to provide information related to production cost generation, and usage for future analytical use cases in manufacturing environment. The research targets to provide a solution for real-time production cost monitoring by utilizing CRISP-DM analytical process and descriptive analytics. The research uses qualitative case study as research methodology and utilizes CRISP-DM process to execute data mining project end to end. The research started by investigating business objectives and business background to discover the business needs, current business processes and organization structure in detail to map out data mining project goals. In latter part of the research CRISP-DM is following to conclude data mining project activities related to data understanding, data cleaning, data modelling and deployment. The results in this research related to each phase of CRISP-DM process by revealing different gaps in target organization’s analytical processes and to provided industry best practices to fill those gaps. The learnings of this research will be exploited in the future data mining projects. Outcome of the research was a descriptive warning system report for users responsible of production cost monitoring for better visibility and decreasing resource used for monitoring. As part of the research and iterative way of working the data mining project increased the knowledge of organization’s cost model, cost generation and cost categorization for business users on manufacturing area.