WIXLIB

IBM SoftwareBusiness AnalyticsIBM SPSS Predictive Operational AnalyticsIBM SPSS predictiveanalytics solutions forwarranty claimsSignificantly reduce warranty costs, improve productquality and enhance customer satisfactionHighlights Determine the root cause of warrantyclaims. Proactively enhance product quality whilereducing warranty costs. Accurately project the amount of warrantyreserve funds required. Identify potentially fraudulent warrantyclaims. Detect and minimize product or deliveryfailures.For many companies, increasing customer expectations combinedwith advances in manufacturing and production technology haveresulted in operational complexity. Manufacturers may lack theplanning capabilities and resources needed to address these issues,which in turn can adversely affect product or delivery quality, leadingto higher warranty costs—and ultimately a greater number ofdissatisfied customers.Warranty-related expenses can amount to anywhere from .5 percentto 7 percent1 of product revenue for large companies, so even smallimprovements in the warranty process can amount to millions ofdollars in savings and lead to significant gains in profitability. Smalland midsize manufacturers can achieve similar results. Manycompanies are also recognizing that their warranty programs canbecome important competitive differentiators that drive customersatisfaction and retention.The warranty analysis process is typically the first post-sale interactionbetween manufacturers and customers, and can be a crucial first stepin improving operational areas such as logistics, quality control andequipment maintenance. Warranty data provides valuable insights thatenable companies to identify the root cause of a claim, such as whetherit is a delivery or production-related issue. Analysis of a firm’s data mayreveal that a defective part is responsible for a high number of warrantyclaims, and that the defect is actually the result of a production qualityissue. Armed with this knowledge, the manufacturer can pinpointthe cause of the production problem and implement an appropriatesolution, eliminating unnecessary claims.

IBM SoftwareBusiness AnalyticsIBM SPSS Predictive Operational AnalyticsPredictiveMaintenanceRaw omerDistributionFigure 1: The typical manufacturing lifecycle consists of multiple steps and touch points. In the automotive industry, for example, a supplier purchases rawmaterials such as oil and leather and manufactures the seats. The seats are delivered to the original equipment manufacturer, which installs them in theautomobile. Finished cars are then placed onto trucks for delivery to retail dealerships. Consumers are a crucial link in this process, because the informationthey provide about defects or delivery issues post purchase enables manufacturers to identify and correct problems at their source.Predictive analytics—the proactiveapproach to warranty managementProactive warranty management requires powerful analyticalcapabilities. Today forward-thinking organizations aredeploying predictive analytics to track and analyze all of thevariables that can affect a product or delivery failure and theassociated warranty costs, so that they can start the problemsolving process earlier than ever before.across the enterprise—including customer information, callcenter and technician notes and sales, performance andproduction data. Using a combination of forecasting, planningand predictive modeling capabilities, the software looks forpatterns of customer complaints and delivery, product or partfailures in real time, and then provides alerts about developingtrends that may indicate a production or performanceproblem to the appropriate decision makers.IBM SPSS Predictive Warranty Analytics providesmanufacturers with the tools to transform warranty claimsdata into actionable intelligence by integrating enormousvolumes of structured and unstructured warranty data fromThe warranty reserve funds of all U.S.-basedmanufacturers were 33.8 billion at the endof 2010.22

IBM SoftwareBusiness AnalyticsIBM SPSS Predictive Operational AnalyticsQuickly find and fix product failuresA premium automobile and motorcycle manufacturer withnumerous production facilities worldwide set out to reduceerror rates and repeat repairs by improving manufacturingand service quality. To accomplish this, the company neededan integrated view of production data, as well as datamanagement and analysis capabilities. IBM SPSS technologyprovided the manufacturer with the tools to proactivelyidentify systematic error patterns and their dependencies. Thecompany was subsequently able to reduce warranty cases from1.1 to 0.85 per vehicle and realized a five percent reduction inwarranty cases, for an annual savings of 30m.With the new emphasis on controlling warranty costs,companies are struggling to understand and correct issues inthe most cost-effective manner. IBM SPSS PredictiveWarranty Analytics helps manufacturers identify a productionproblem early on by determining the root cause, calculatinghow much it will cost to fix, and then taking the appropriatesteps to correct it.When an automobile manufacturer learned that side mirrorswere repeatedly breaking off, it turned to IBM SPSSwarranty analytics solutions to resolve the issue. Using thesoftware to analyze structured and text data from dealerships,the corporate website and social media sites, the companyfound that a specific machine used to manufacture the mirrorsub-assembly had not been maintained properly and wasproducing low-quality parts. By making operational adjustments, the company eliminated the risk of producing futuredefective parts that could increase warranty exposure. What’smore, the company was able to notify customers who werelikely to experience the failure but who had not brought theirvehicles in for service yet, improving customer satisfaction.Detect fraud early in the warrantylifecycleAnother important benefit of IBM SPSS warranty analyticstechnology is that it helps manufacturers detect andminimize warranty claim fraud. In many cases of fraud,anomalies can be seen in data. But given the steadilyincreasing volume of warranty information, subtle patternsmay be easily overlooked.IBM SPSS solutions help mitigate losses from warranty fraudby enabling organizations to more quickly identify suspiciousbehavior within large amounts of data. By combiningpowerful analytical techniques with knowledge of the warranty process, manufacturing companies can recognize thecharacteristics or inputs that tend to increase warranty fraud,and use these insights to create an appropriate interventionstrategy for each probable instance or type of fraud.In addition to ensuring product quality, IBM SPSS warrantyanalytics solutions can help reduce the costs associated withproduction line maintenance. By using predictive analytics toanalyze machinery inputs and outputs, manufacturers canforecast when certain components or devices may break downor need maintenance, so that product quality is not affectedand warranty claims can be minimized.For example, an industrial products manufacturer usingIBM SPSS warranty analytics software may notice asignificant number of returns during the last week of eachmonth for three consecutive months. Using the software’sanomaly detection capabilities, the company could analyzethe quality technician’s logs in conjunction with productionline information. This information may reveal that theparts in question left the production facility meeting allspecifications, indicating that the returned parts either werenot manufactured by the company or were damaged duringthe delivery process.Toyota Motor Corp. paid approximately 4.2 billion in warranty claims in 2010.33

IBM SoftwareBusiness AnalyticsIBM SPSS Predictive Operational AnalyticsDefine optimal warranty terms andconditionsThe company relies on IBM SPSS software to analyzedata on vehicles and repairs, vehicle error memories anddealer feedback in combination with other information.What sets this solution apart is the fact that data is nolonger considered in isolation, but in combination,providing completely new insights.Deciding what type of warranty policies to offer can bechallenging for manufacturers, who need to create costeffective policies that also satisfy customer expectations. Apoor warranty policy can result in financial losses or evenimply inferior product quality.The results of the analyses and customer feedback areimmediately channeled back into BMW’s working processesand distributed to a wide circle of users, helping theorganization reduce error rates, save costs and improvecustomer satisfaction.The warranty period for automobiles was90 days in the 1930s and this has steadilyincreased over the years to 10-year warrantyterms for certain models and parts.4IBM SPSS Predictive Warranty Analytics can helporganizations determine the optimal warranty term, basedon cost, improving terms and conditions for all parties:customers, suppliers and manufacturers. By enabling firmsto evaluate data on past and present claims, warrantywriters can determine the most cost-efficient service andreplacement guidelines.Business BenefitsIBM is the leader in predictive analytics solutions thathelp manufacturers: Identify and resolve delivery and quality issues earlier inthe product lifecycle, leading to fewer warranty claims. Detect patterns of fraudulent warranty claim behavior anddevelop appropriate policies and controls.Include warranty programs in customersatisfaction and retention strategies Improve customer satisfaction by offering more reliableproducts and more attractive warranty terms.In a fiercely competitive global marketplace, many manufacturers are beginning to view their warranty programs asmore than just repairing or replacing defective products.They are recognizing that these programs can be valuableopportunities to build customer loyalty and enhancebrand reputation. Strengthen competitive advantage through more positivebrand image, higher service levels and lower total cost ofownership (TCO).As a premium manufacturer, The BMW Group aims towin customers over with innovative, original designs andquality, so it is essential to continually evaluate and assessits products and services and take customers’ opinionson board.4