Survey into predictive key performance indicator analysis from data mining perspective

Thakur, Akshay; Beck, Robert E.; Mostaghim, Sanaz; Grosmann, Daniel

doi:10.1109/etfa46521.2020.9212111

Cited by 7 publications

(7 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Apart from CRISP-DM, there are similar data mining methodologies like SEMMA (Sample, Explore, Modify, Model & Assess), DMAIC (Define, Measure, Analyze, Improve& Control) etc. As per the 2007 survey (167 respondents) who have implemented predictive analyses, only 15% used CRISP-DM, and the majority were found using their own methodology (52%) [4] [8] [9].…”

Section: Predictive Analytics For Kpimentioning

confidence: 99%

Prediction of Top Key Performance Indicator in Automotive Production System using Data Mining

2022

Proceedings of WCSE 2022 Spring Event: 2022 9th International Conference on Industrial Engineering and Applications

View full text Add to dashboard Cite

In the past years, the Automotive industry has faced many challenges such as shifting market, digitalization, increased competition, to recent issues of semiconductor shortages. Technology is transforming the automotive industry at a fast pace, and predictive analytics is at its core if harnessed efficiently. Predictive analytics present significant opportunities by using techniques like data mining and machine learning by aiding decision-makers to optimize production processes in the manufacturing plants. Forecasting the critical key performance indicators also ensures a decision support system for the domain experts to take corrective actions. In this research paper, a case study is carried out at the Volkswagen passenger car manufacturing plant in Germany with three years of data to predict one of the top key performance indicators (HPV -Hours Per Vehicle) from the production system. In the global automotive industry, HPV is considered a dominant controlling indicator in production systems. Predicting this HPV ensures robust production planning and provides transparency about the influencing variables so that necessary measures can be taken proactively to improve HPV compared to the planned HPV budget. Comparison between different machine learning algorithms such as Decision tree, Neural network, Linear Regression etc., is done to find accurate machine learning models for key performance indicator prediction based on historical data. Case study results indicate that a Neural network can predict HPV with 2.5% relative error based on historical data. A higher coefficient of determination (R 2 =0.8) illustrates that the selected model is stable for prediction. The results show that the machine learning algorithms can be effectively used for forecasting the HPV for the Volkswagen plant and understanding the impact of influencing factors on HPV.

show abstract

Section: Predictive Analytics For Kpimentioning

confidence: 99%

Prediction of Top Key Performance Indicator in Automotive Production System using Data Mining

2022

Proceedings of WCSE 2022 Spring Event: 2022 9th International Conference on Industrial Engineering and Applications

View full text Add to dashboard Cite

show abstract

“…Economic Success Criteria: It is best practice in business monitoring [46] and manufacturing [47] to add an economic success criterion in the form of a Key Performance Indicator (KPI) to the project. A KPI is an economic measure for the current and future business relevance of an application and requires a precise definition ( [48], ISO 22400).…”

Section: Resource Demandmentioning

confidence: 99%

“…A KPI is an economic measure for the current and future business relevance of an application and requires a precise definition ( [48], ISO 22400). Adding a KPI to the machine learning application helps to objectify the business goals of the ML application and can be used for decision making [46]. Once the application is deployed, predictions of a future KPI based on past and present data is applicable [46] and, e.g., costs can be weighted by their expected probability of occurrence [49].…”

Section: Resource Demandmentioning

confidence: 99%

“…Adding a KPI to the machine learning application helps to objectify the business goals of the ML application and can be used for decision making [46]. Once the application is deployed, predictions of a future KPI based on past and present data is applicable [46] and, e.g., costs can be weighted by their expected probability of occurrence [49]. Deviations from the defined success criteria are an indicator for risk.…”

Section: Resource Demandmentioning

confidence: 99%

“…An equal approach can be applied to the outputs of the model that underlie a certain distribution if environment conditions are stable and can give an estimate on the number of actions performed when interacting with an environment [36]. The monitoring phase also includes a comparison of the current performance or a predicted future performance [46] with the defined success criteria. The future performance of the application can be predicted using ML.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Towards CRISP-ML(Q): A Machine Learning Process Model with Quality Assurance Methodology

Studer

Bui

Drescher

et al. 2021

MAKE

129

View full text Add to dashboard Cite

Machine learning is an established and frequently used technique in industry and academia, but a standard process model to improve success and efficiency of machine learning applications is still missing. Project organizations and machine learning practitioners face manifold challenges and risks when developing machine learning applications and have a need for guidance to meet business expectations. This paper therefore proposes a process model for the development of machine learning applications, covering six phases from defining the scope to maintaining the deployed machine learning application. Business and data understanding are executed simultaneously in the first phase, as both have considerable impact on the feasibility of the project. The next phases are comprised of data preparation, modeling, evaluation, and deployment. Special focus is applied to the last phase, as a model running in changing real-time environments requires close monitoring and maintenance to reduce the risk of performance degradation over time. With each task of the process, this work proposes quality assurance methodology that is suitable to address challenges in machine learning development that are identified in the form of risks. The methodology is drawn from practical experience and scientific literature, and has proven to be general and stable. The process model expands on CRISP-DM, a data mining process model that enjoys strong industry support, but fails to address machine learning specific tasks. The presented work proposes an industry- and application-neutral process model tailored for machine learning applications with a focus on technical tasks for quality assurance.

show abstract

Digital Age, Business Strategy, and Firm’s Performance

2022

View full text Add to dashboard Cite

Survey into predictive key performance indicator analysis from data mining perspective

Cited by 7 publications

References 24 publications

Prediction of Top Key Performance Indicator in Automotive Production System using Data Mining

Prediction of Top Key Performance Indicator in Automotive Production System using Data Mining

Towards CRISP-ML(Q): A Machine Learning Process Model with Quality Assurance Methodology

Digital Age, Business Strategy, and Firm’s Performance

Contact Info

Product

Resources

About