Human Error Prediction and Control Model Using Recursive Partitioning
Publication Type:Conference Paper
Source:Proceedings of the 2014 International Conference on Interdisciplinary Advances in Applied Computing, ACM, Amrita University (2014)
Keywords:count data, error management, Human error, recursive partitioning, transaction processing
The problems associated with human error are very complicated with a number of dynamic factors influencing the outcome. Though it has been studied in detail in various industries with various tools and techniques, there is no comprehensive model available considering multiple factors that address the issue. In the field of Banking and Financial services, the problem of human error is more critical since it can lead to operational losses and bad customer experience. Traditionally simple parametric and non - parametric statistical tests of hypotheses are used as standalone tools for analysis and hence improvement. From a real world perspective, control of one factor leading to a trade-off on another resulting in more improvement projects rather than resolving the problems. The specific requirement from a practitioners point of view is also not just identifying factors influencing errors but to what extent. Poisson regression, Negative Binomial regression and Recursive Partitioning are the tools which help us analyze the problem irrespective of the type of data and arrive at controllable thresholds for the Operations Managers. This paper aims at providing a comprehensive and practical error management and control model for human errors in a transaction processing industry.
Cite this Research Publication
P. Balasubramanian and Kalyanasundaram, K., “Human Error Prediction and Control Model Using Recursive Partitioning”, in Proceedings of the 2014 International Conference on Interdisciplinary Advances in Applied Computing, Amrita University, 2014.
Related Research Publications
- Effect of Spirituality on Human Performance-A Myth or Reality?
- Predicting remaining useful life of cutting tools with regression and ANN analysis
- Modeling Potential-Adopter Student Acceptance
- Enhancing design space exploration by extending CPU/GPU specifications onto FPGAs
- Review of efficient and innovative mathematical models on VLSI circuit partitioning