Accurately predicting insurance claims is crucial for insurance companies as it directly impacts cash flow, pricing strategies and overall profitability. This paper addresses the critical need for distributional forecasting, particularly for regulatory requirements that necessitate understanding the variability of the claims. To this end, a hybrid model is proposed that integrates the chain ladder (CL) method, machine learning (ML) and fuzzy logic for enhanced claims reserve prediction through three major modules. In the first module, a multi-layer perceptron (MLP) is employed to capture patterns of the losses’ location, and then, CL estimations are imported to a second module using the adaptive neuro-fuzzy inference system (ANFIS). The outputs of the MLP and ANFIS are integrated into a recurrent neural network (RNN), as a kind of ensemble learning approach, in the third module. Specifically, a long short-term memory (LSTM) model is employed to obtain the mean and standard deviation of the loss distribution for each cell in the claims development triangle. Moreover, a module-wise explainability strategy is adopted that combines intrinsically interpretable components with post-hoc analyses for neural networks. Through comprehensive simulations, parameter optimization is addressed, the importance of its key components is validated, and the competitive performance of the method in accurately predicting the outstanding loss liabilities is demonstrated.

In the area of multivariate process quality control, it is sometimes important to monitor the ratio of two normal random variables denoted by RZ over time. The concept of control charts has often been harnessed in this field, leading to the application of various types of statistical models, including Shewhart, Exponentially Weighted Moving Average (EWMA), and so forth. However, there is little attention to implementation of machine learning-based control charts. To bridge this gap, a novel machine learning based model incorporating the attention mechanism approach, as an implemented Artificial Intelligence (AI) model, is proposed to monitor the RZ in Phase II applications. The proposed RZ method not only provides quicker Out-of-Control (OC) shift detection than conventional RZ control charts but also does not require the quality controller to have any prior information about the upward or downward shift patterns, which is a major assumption in most of the previous RZ models. We provide extensive performance comparison results to discuss the statistical performance of our proposed method through Monte Carlo simulations. Moreover, a comprehensive real example about surveillance of the cryptocurrency market is provided to illustrate the practical application of our proposed method. Through simulation and back-testing results, it is shown how the proposed method can lead to an automated trading strategy.

Monitoring healthcare processes, such as surgical outcomes, with a keen focus on detecting changes and unnatural conditions at an early stage is crucial for healthcare professionals and administrators. In line with this goal, control charts, which are the most popular tool in the field of Statistical Process Monitoring, are widely employed to monitor therapeutic processes. Healthcare processes are often characterized by a multistage structure in which several components, states or stages form the final products or outcomes. In such complex scenarios, Multistage Process Monitoring (MPM) techniques become invaluable for monitoring distinct states of the process over time. However, the healthcare sector has seen limited studies employing MPM. This study aims to fill this gap by developing an MPM control chart tailored for healthcare data to promote early detection, confirmation, and patient safety. As it is important to detect unnatural conditions in healthcare processes at an early stage, the statistical control charts are combined with machine learning techniques (i.e., we deal with Intelligent Control Charting, ICC) to enhance detection ability. Through Monte Carlo simulations, our method demonstrates better performance compared to its statistical counterparts. To underline the practical application of the proposed ICC framework, real data from a two-stage thyroid cancer surgery is utilized. This real-world case serves as a compelling illustration of the effectiveness of the developed MPM control chart in a healthcare setting.

The widespread implementation of computer technology has led to an increased use of Machine Vision Systems (MVS) for quality control in advanced manufacturing industries. The application of Statistical Process Control (SPC) techniques, especially control charts, is one of the most important and effective ways for fault detection in this field. Several statistical control charts have been extended to monitor images in the framework of MVS where the major aim is to improve the on-line detection ability, which is equivalent to avoid production of nonconforming products. In various industrial and non-industrial applications, control charts have been shown to be better able to detect anomalies in the underlying process(es) by integrating machine and ensemble learning techniques. However, in the field of image monitoring, approaches based on intelligent techniques have rarely been proposed. To bridge this gap and to improve the detection ability of control charts in monitoring image-based quality characteristics, this paper presents a novel control chart that combines machine learning, ensemble learning and image partitioning. In order to reach the maximum performance in Phase II SPC applications, specific designing and parameter tuning procedures are provided. The superiority of the proposed Partitioning Ensemble Control Chart (PECC) is verified by comparative analysis including conventional monitoring schemes by means of extensive simulations, in which different fault sizes and fault locations are applied to the images. Finally, a real MVS for quality control of O-rings is discussed to illustrate the practical application of the PECC.

Abstract Many researchers have shown interest in profile monitoring; however, most of the applications in this field of research are developed under the assumption of normal response variable. Little attention has been given to profile monitoring with non-normal response variables, known as general linear models which consists of two main categories (i.e., logistic and Poisson profiles). This paper aims to monitor Poisson profile monitoring problem in Phase II and develops a new robust control chart using support vector regression by incorporating some novel input features and evolutionary training algorithm. The new method is quicker in detecting out-of-control signals as compared to conventional statistical methods. Moreover, the performance of the proposed scheme is further investigated for Poisson profiles with both fixed and random explanatory variables as well as non-parametric profiles. The proposed monitoring scheme is revealed to be superior to its counterparts, including the likelihood ratio test (LRT), multivariate exponentially weighted moving average (MEWMA), LRT-EWMA and other machine learning-based schemes. The simulation results show superiority of the proposed method in profiles with fixed explanatory variables and non-parametric models in nearly all situations while it is not able to be the best in all the simulations when there are with random explanatory variables. A diagnostic method with machine learning approach is also used to identify the parameters of change in the profile. It is shown that the proposed profile diagnosis approach is able to reach acceptable results in comparison with other competitors. A real-life example in monitoring Poisson profiles is also provided to illustrate the implementation of the proposed charting scheme.

Abstract In profile monitoring, it is usually assumed that the observations between or within each profile are independent of each other. However, this assumption is often violated in manufacturing practice, and it is of utmost importance to carefully consider autocorrelation effects in the underlying models for profile monitoring. For this reason, various statistical control charts have been proposed to monitor profiles when between- or within-data is correlated in Phase II, in which the main aim is to develop control charts with quicker detection ability. As a novel approach, this study aims to employ machine learning techniques as control charts instead of statistical approaches in monitoring profiles with between-profile autocorrelations. Specifically, new input features based on conventional statistical control chart statistics and normalized estimated parameters are defined that are capable of adequately accounting for the between-autocorrelation effect of profiles. In addition, six machine learning techniques are extended and compared by means of Monte Carlo simulations. The simulation results indicate that machine learning techniques can obtain more accurate results compared with statistical control charts. Moreover, adaptive neuro-fuzzy inference systems outperform other machine learning techniques and the conventional statistical control charts.