Warranty Claims Data Research Articles

A finite mixture model for multiple dependent competing risks with applications of automotive warranty claims data

A finite mixture model for multiple dependent competing risks with applications of automotive warranty claims data

Dynamic monitoring of polynomial profiles with attribute responses and between-profile correlation

Profile monitoring is a popular statistical process control problem in recent years. In many applications, quality characteristics of interest are attribute data due to the inherent feature of processes or limitation of data collection costs. The correlation among data is becoming more significant, since data collection intervals are becoming shorter in the big data era. However, research on monitoring profiles with attribute responses in the presence of Between-Profile Correlation (BPC) has received relatively scant attention. Motivated by a real example of automobile warranty claims, this article aims to monitor polynomial profiles with attribute responses and BPC. The generalized polynomial model and the learning curve model are adopted to characterize the profile relationship and the correlation between profiles, respectively. Then, an EWMA chart with dynamic control limits (dEWMA) is developed. Simulation studies show that ignoring the BPC does not affect the in-control performance of the chart with dynamic control limits, but does have devastating effects on the out-of-control performance. The proposed dEWMA chart can address the impact of correlation and provide superior monitoring performance compared with some competitors. Finally, a real example of warranty claims data is presented to illustrate the implementation of the proposed chart.

A Snapshot-Stacked Ensemble and Optimization Approach for Vehicle Breakdown Prediction

Predicting breakdowns is becoming one of the main goals for vehicle manufacturers so as to better allocate resources, and to reduce costs and safety issues. At the core of the utilization of vehicle sensors is the fact that early detection of anomalies facilitates the prediction of potential breakdown issues, which, if otherwise undetected, could lead to breakdowns and warranty claims. However, the making of such predictions is too complex a challenge to solve using simple predictive models. The strength of heuristic optimization techniques in solving np-hard problems, and the recent success of ensemble approaches to various modeling problems, motivated us to investigate a hybrid optimization- and ensemble-based approach to tackle the complex task. In this study, we propose a snapshot-stacked ensemble deep neural network (SSED) approach to predict vehicle claims (in this study, we refer to a claim as being a breakdown or a fault) by considering vehicle operational life records. The approach includes three main modules: Data pre-processing, Dimensionality Reduction, and Ensemble Learning. The first module is developed to run a set of practices to integrate various sources of data, extract hidden information and segment the data into different time windows. In the second module, the most informative measurements to represent vehicle usage are selected through an adapted heuristic optimization approach. Finally, in the last module, the ensemble machine learning approach utilizes the selected measurements to map the vehicle usage to the breakdowns for the prediction. The proposed approach integrates, and uses, the following two sources of data, collected from thousands of heavy-duty trucks: Logged Vehicle Data (LVD) and Warranty Claim Data (WCD). The experimental results confirm the proposed system’s effectiveness in predicting vehicle breakdowns. By adapting the optimization and snapshot-stacked ensemble deep networks, we demonstrate how sensor data, in the form of vehicle usage history, contributes to claim predictions. The experimental evaluation of the system on other application domains also indicated the generality of the proposed approach.

Monitoring aggregate warranty claims with dynamically designed CUSUM and EWMA charts

Statistical monitoring of warranty claims data using dynamic probability control limits has been shown to be effective in early detection of unforeseen reliability problems that emerge at the design and manufacturing phases. As the discrepancy between abnormal patterns and the normal pattern in aggregate warranty claims is usually small (especially at the early stage), we develop two new dynamic monitoring schemes that adopt CUSUM-type and EWMA-type statistics, named DyCUSUM and DyEWMA, respectively, to better address the warranty claims monitoring problem. Three effective algorithms – that is, the Monte Carlo simulation, Markov chain, and near-enumeration algorithms – are proposed to progressively determine control limits for the two schemes. In particular, comparison studies show that the near-enumeration algorithm can attain a higher approximation accuracy with a lower computational burden and is thus recommended. In-depth simulation experiments are then conducted to assess the performance of the schemes. We find that the DyEWMA scheme has superior and robust detection performance in various situations, whereas the DyCUSUM scheme is less effective and could even be ineffective in certain cases, compared with a Shewhart-type counterpart. Some specific suggestions are also provided to facilitate implementation of the proposed monitoring schemes. Improved schemes by combining the moving window approach to mitigate the ‘inertia’ problem is further discussed. Finally, a real case study is presented.

Clustered Coefficient Regression Models for Poisson Process with an Application to Seasonal Warranty Claim Data

Motivated by a product warranty claims dataset, we propose clustered coefficient regression models in a nonhomogeneous Poisson process for recurrent event data. The proposed method, referred as CLUPP, can estimate the group structure and parameters simultaneously. In our proposed method, a penalized regression approach is used to identify the group structure. Numerical studies show that the proposed approach can identify the group structure well, and outperforms traditional methods such as hierarchical clustering and K-means. We also establish theoretical properties, which show that the proposed estimators can converge to true parameters in high probability. In the end, we apply our proposed methods to the product warranty claims dataset, which achieve better prediction than the state-of-the-art methods.

Warranty Reserve Management: Demand Learning and Funds Pooling

Problem definition: Warranty reserves are funds used to fulfill future warranty obligations for a product. In this paper, we investigate the warranty reserve planning problem faced by a manufacturing firm who manages warranties for multiple products. Academic/practical relevance: It is nontrivial to determine a proper amount of reserves to hold, because warranty expenditures are random in nature and reserving either excess or insufficient cash would incur losses. How can warranty reserve levels be optimized and promptly adjusted is a focal issue, especially for firms selling multiple products. Methodology: Inspired by the general pattern of empirical warranty claims data, we first develop an aggregate warranty cost (AWC) forecasting model for a single product by coupling stochastic product sales and failure processes, which can be used to plan for warranty reserves periodically. The reserve levels are then optimized via a distributionally robust approach, because the exact distribution of AWC is generally unknown. To reduce the losses generated from managing the funds, we further investigate two potential loss-reduction approaches: demand learning and funds pooling. Results: For the demand learning algorithm, we prove that, as the sales period grows, the optimal learning parameter asymptotically converges to a constant in a fairly fast rate; our simulation experiments show that the performance of demand learning is promising and robust under general warranty claim patterns. Moreover, we find that the benefits of funds pooling change over different stages of the warranty life cycle; in particular, the relative pooling benefit in terms of reserve losses is nonincreasing over time. Managerial implications: This study offers guidelines on how manufacturers should adaptively forecast and dynamically plan warranty reserves over the warranty life cycle.

Designing differential service strategy for two-dimensional warranty based on warranty claim data under consumer-side modularisation

With the development of modular manufacturing technology and diversified demand for products, the demand for differential warranty service has become increasingly prominent. This article proposes a differential service strategy for two-dimensional warranty using warranty claim data under consumer-side modularisation. In contrast to previous research that optimised maintenance strategies under a given failure density function, this article obtains actual usage rate and failure density functions of main failure types by determining the function type and parameter fitting. A detailed case study of a product under a two-dimensional warranty is discussed to interpret the proposed model. Using the mathematical model proposed in this research, warranty service providers can offer consumers a differential warranty service under consumer-side modularisation.

Warranty Claims Forecasting for New Products Sold with a Two-Dimensional Warranty

Warranty claims forecasting plays an increasingly important role not only for preparing financial plans but also for optimizing warranty policy and improving after-sale services. In the case of new products, an important feature is that the new generation of products often has a close connection with the previous generations of products it replaces. Thus, the warranty claims data of the previous generations of products can be used for extracting reliability information of new products. In this context, we propose a warranty claims forecasting model considering usage rate for new products sold with a two-dimensional warranty. The accelerate failure time model is introduced to investigate the effect of usage rate on product degradation. The non-homogeneous Poisson process is used to model failure counts of repairable products and the constrained maximum likelihood estimation method is used to estimate model parameters. The results of data experiments based on both simulation and real data collected from an automobile manufacturer in China show that the proposed model considering the varying usage rate outperforms the traditional models in forecasting the number of warranty claims.

On dynamically monitoring aggregate warranty claims for early detection of reliability problems

Warranty databases managed by most world-leading manufacturers are constantly expanding in the big data era. An important application of warranty databases is to detect unobservable reliability problems that emerge at design and/or manufacturing stages, through modeling and analysis of warranty claims data. Usually, serious reliability problems will result in certain abnormal patterns in warranty claims, which can be captured by appropriate statistical methods. In this article, a dynamic control charting scheme is developed for early detection of reliability problems by monitoring warranty claims one period after another, over the product life cycle. Instead of specifying a constant control limit, we determine the control limits progressively by considering stochastic product sales and non-homogeneous failure processes, simultaneously. The false alarm rate at each time period is controlled at a desired level, based on which abrupt changes in field reliability, if any, will be detected in a timely manner. Furthermore, a maximum-likelihood-based post-signal diagnosis scheme is presented to aid in identifying the most probable time of problem occurrence (i.e., change point). It is shown, through in-depth simulation studies and a real case study, that the proposed scheme is able to detect an underlying reliability problem promptly and meanwhile estimate the change point with an acceptable accuracy. Finally, a moving window approach concerning only recent production periods is introduced to extend the original model so as to mitigate the “inertia” problem.

Seasonal warranty prediction based on recurrent event data

Warranty return data from repairable systems, such as home appliances, lawn mowers, computers and automobiles, result in recurrent event data. The nonhomogeneous Poisson process (NHPP) model is used widely to describe such data. Seasonality in the repair frequencies and other variabilities, however, complicate the modeling of recurrent event data. Not much work has been done to address the seasonality, and this paper provides a general approach for the application of NHPP models with dynamic covariates to predict seasonal warranty returns. The methods presented here, however, can be applied to other applications that result in seasonal recurrent event data. A hierarchical clustering method is used to stratify the population into groups that are more homogeneous than the overall population. The stratification facilitates modeling the recurrent event data with both time-varying and time-constant covariates. We demonstrate and validate the models using warranty claims data for two different types of products. The results show that our approach provides important improvements in the predictive power of monthly events compared with models that do not take the seasonality and covariates into account.

Investigating the effects of operators on warranty cost under sales delay conditions

PurposeDue to the fact that the non-standard products, being used by customers, may cause failures in products with sales delays, which naturally affect the warranty policy. Thus, it seems to be necessary to study these two concepts simultaneously. The paper aims to discuss these issues.Design/methodology/approachIn this paper, a model is developed for estimating the expected warranty costs under sales delay conditions when two operator costs (failing but not reported and non-failing but reported) are included.FindingsThe proposed model is validated using a numerical example for a two types of intermittent and fatal failures occur under a non-renewing warranty policy.Originality/valueSales delay is the time interval between the date of production and the date of sale. Most reported literature on warranty claims data analysis related to sales delay have mainly focussed on estimating the probability distribution of the sales delay.

Karta kontrolna do wieloetapowego monitorowania produktów sprzedawanych z gwarancją dwuwymiarową, z korektą ryzyka i wczesne ostrzeganie o wadach produkcyjnych na podstawie danych z reklamacji

Warranty claims data contain valuable information about the quality and reliability of products. The monitoring and early-warning of warranty claims data are of great significance to the manufacturer by identifying and solving the emerging quality or reliability problem as soon as possible. However, though it has been used widely in the automobile industry, there are no studies that have been carried out on the monitoring and early-warning of claims data for products sold with two-dimensional warranty. In order to fill this gap, fitting the two-dimensional warranty claims data with accelerated failure model (AFT), a multi-stage riskadjusted control chart is proposed by this paper, for which a reasonable product sales tracking time and a monitoring time are suggested to reduce the influence of sales delay and fluctuating claim rates. Comparing with traditional Cumulative Sum Control Chart (CUSUM), the applicability and availability of the proposed model are demonstrated in the final.

Warranty cost analysis: Increasing warranty repair times

In this study, we model the warranty servicing costs under nonrenewing and renewing free repair warranties. We assume nonzero increasing repair times with the warranty cost depending on the length of the repair time. An increasing geometric process is used to model the consecutive repair times. We introduce the generalised alternating renewal process, which is an alternating process with cycles consisting of an item's operational time followed by the corresponding repair time. We derive analytical results for a generalised alternating renewal process with a finite time horizon and use them to evaluate the warranty costs over the warranty period and over the life cycle of the product under the nonrenewing free repair warranty and renewing free repair warranty. Properties of the model are demonstrated through a simulation study and through the application to warranty claims data from an automotive manufacturer.

Doubly geometric processes and applications

The geometric process has attracted extensive research attention from authors in reliability mathematics since its introduction. However, it possesses some limitations, which include that: (1) it can merely model stochastically increasing or decreasing inter-arrival times of recurrent event processes and (2) it cannot model recurrent event processes where the inter-arrival time distributions have varying shape parameters. Those limitations may prevent it from a wider application in the real world. In this paper, we extend the geometric process to a new process, the doubly geometric process, which overcomes the above two limitations. Probability properties are derived, and two methods of parameter estimation are given. Application of the proposed model is presented: one is on fitting warranty claim data, and the other is to compare the performance of the doubly geometric process with the performance of other widely used models in fitting real-world datasets, based on the corrected Akaike information criterion.

Bayesian approach to hazard rate models for early detection of warranty and reliability problems using upstream supply chain information

Hazard rate models are proposed recently for detection of reliability problems using information from upstream supply chain and warranty databases. Whereas these models improve the accuracy of reliability problem detection, they require relatively long lead-times due to their reliance on just the actual warranty claims data collected from the field. We propose a Bayesian approach to hazard rate models that reduces the need for extensive warranty claim history. The paper introduces Bayesian hazard rate models to account for uncertainties of the explanatory covariates, in particular, information collected during product development, major design change/upgrade efforts, and manufacturing technology upgrades. In doing so, it improves both the accuracy of extant hazard rate models for reliability problem detection as well as the lead-time for detection. The proposed methodology is illustrated and validated using real-world data from a leading global Tier-1 automotive supplier.

Analysis of Field Return Data With Failed-But-Not-Reported Events

ABSTRACTWarranty data contain valuable information on product field reliability and customer behaviors. Most previous studies on analysis of warranty data implicitly assume that all failures within the warranty period are reported and recorded. However, the failed-but-not-reported (FBNR) phenomenon is quite common for a product whose price is not very high. Ignorance of the FBNR phenomenon leads to an overestimate of product reliability based on field return data or an overestimate of warranty cost based on lab data or tracking data. Being an indicator of customer satisfaction, the FBNR proportion provides valuable managerial insights. In this study, statistical inference for the FBNR phenomenon as well as field lifetime distribution is described. We first propose a flexible FBNR function to model the time-dependent FBNR behavior. Then, a framework for data analysis is developed. In the framework, both semiparametric and parametric approaches are used to jointly analyze warranty claim data and supplementary tracking data from a follow-up of selected customers. The FBNR problem in the tracking data is minimal and thus the data can be used to effectively decouple the FBNR information from the warranty claim data. The proposed methods are illustrated with an example. Supplementary materials for this article are available online.

Some reliability issues for incomplete two-dimensional warranty claims data

Bivariate reliability and vector bivariate hazard rate or hazard gradient functions are expected to have a role for meaningful assessment of the field performance for items under two-dimensional warranty coverage. In this paper a usage rate based simple class of bivariate reliability function is proposed and various bivariate reliability characteristics are studied for warranty claims data. The utilities of such study are explored with the help of a real life synthetic data.

Age-based analysis of automobile component failure data: a case study

This paper presents techniques for the age-based analysis of failure data of an automobile component. An automobile component may fail for different failure modes. It is important to identify the failure mode for which the automobile component fails more frequently. To identify important failure modes, this paper first presents a Pareto chart and preliminary analysis of warranty claims data. Then it estimates the warranty claims rate WCR, nonparametric cumulative density function CDF and parametric CDF for different failure modes. These estimates can be used to assess and compare the lifetime distributions and reliability functions of the component for different failure modes. The paper also discusses a method for estimating the reliability of the component by eliminating the dominant failure modes. These results can be used to investigate how the reliability of the component improves by removing a failure mode. The results of the paper would be useful to reliability engineers, managers, practitioners, and customers for cost-benefit analysis, including improvement in reliability, reduction in warranty cost, and predicting claims rates and costs.

Age-based analysis of automobile component failure data: a case study

This paper presents techniques for the age-based analysis of failure data of an automobile component. An automobile component may fail for different failure modes. It is important to identify the failure mode for which the automobile component fails more frequently. To identify important failure modes, this paper first presents a Pareto chart and preliminary analysis of warranty claims data. Then it estimates the warranty claims rate (WCR), nonparametric cumulative density function (CDF) and parametric CDF for different failure modes. These estimates can be used to assess and compare the lifetime distributions and reliability functions of the component for different failure modes. The paper also discusses a method for estimating the reliability of the component by eliminating the dominant failure modes. These results can be used to investigate how the reliability of the component improves by removing a failure mode. The results of the paper would be useful to reliability engineers, managers, practitioners, and customers for cost-benefit analysis, including improvement in reliability, reduction in warranty cost, and predicting claims rates and costs.

Modelling an Optimized Warranty Analysis methodology for fleet industry using data mining clustering methodologies with Fraud detection mechanism using pattern recognition on hybrid analytic approach

In this paper we have analyzed the huge volume of warranty data for segregating the fraudulent warranty claims using pattern recognition and clustering methodology. Recent survey of automotive industry shows up to 10% of warranty costs are related to warranty claims fraud, costing manufacturers several billions of dollars. Most of the automotive companies are suspecting and aware of warranty fraud. But they are not sure of the extent and ways to eliminate it. The existing methods to detect warranty fraud are very complex and expensive as they are dealing with inaccurate and vague data, causing manufacturers to bear the excessive costs. We are proposing model to find anomalies on warranty data along with component failure data and patterns based on historic warranty claims data under particular region and for specific component as the data are of high volume. We are managing to isolate all the imapcting the factors that indicate a claim, that has a high probability of fraudulence such as failure date and claim date, mode of failure etc., In addition to this we discover suspecting claims that have the greatest adjustment potential for further review by claim process. We altogether integrating data with with claims processing, reports and business rules along with reported mode of failure as we are minimizing changes to existing systems, since the analysis is carried out by identifying patterns. Since we are working with factual data, it gives more room to identify the actual cost involved on warranty claim.