Traditionally, companies used to take up to six months to be able to deploy their predictive analytic models. In this scenario, whenever the data scientist team finished building the best model out of historical data, it had to be custom coded into production by the IT Engineering team. This lengthy process has no place in the big data era, where data is being generated fast and changing rapidly. Fortunately PMML, the Predictive Model Markup Language, turned the traditional model deployment process upside-down. With PMML, there is no need for recoding of a predictive solution. As a consequence, models are deployed in minutes, not months.
Zementis offers two PMML-based scoring platforms. ADAPA® for real-time and UPPI® for big data scoring. By leveraging the power of ADAPA and UPPI, users benefit from predictive solutions ranging from forecasting customer behavior and risk to the prediction of failures in machinery. Once developed, predictive solutions need to be deployed in an agile and consistent way. ADAPA and UPPI offer the conduit for predictive analytics to work its magic: crack the big data code.
Financial institutions are relying more and more on on-line transactions, an area in which fraud and abuse need to be tackled in real-time. With ADAPA, companies combine the power of predictive analytics to take proactive actions to stop abuse as soon as risky patterns are detected. Although real-time analytics provides a vital resource against fraud and abuse, big data analytics can provide a much larger context into any prediction. With UPPI for big data scoring, financial institutions are able to establish connections between a number of disparate data sources and detect fraud patterns not previously seen.
ADAPA® is the perfect conduit for predictive marketing and on-line advertising since it provides the back-bone for efficient deployment of targeted campaigns and strategies in a multi-channel environment in which internet-enabled mobile devices are becoming ubiquitous. With ADAPA, companies benefit from a consistent and scalable platform to decide in real-time on the right marketing strategy across different geographies, demographics, psychographics and channels. With UPPI, on the other hand, companies are able to score vast amounts of data against one or many models and use the results to drive their marketing strategy in its entirety.
In the global quest for Smart Grid technology, predictive analytics enables intelligent solutions which save energy, reduce cost, and increase reliability. Zementis has joined forces with SAIC, a FORTUNE 500® scientific, engineering, and technology applications company, to deliver advanced predictive analytic solutions to utility companies. With ADAPA®, companies monitor and analyze in real-time all grid components, including smart meters and transformers. With UPPI for big data scoring, data and models benefit from scoring in-database or Hadoop in a scale never seen before.
Quality Control Solutions
Predictive analytics is revolutionizing quality control in production lines around the globe, since it delivers a consistent and deterministic approach for the detection of problems. ADAPA and UPPI are ideal for quality control in which manufactured parts need to checked for defects before they are packaged and sent away or used as part of a bigger piece of machinery. By integrating defect-detection into their business decisions, companies are able to decide instantly on next steps as well as track down the source of a defect and correct it accordingly, fast.
Predictive Maintenance Solutions
In today’s factories, the breakdown of an essential piece of machinery can directly undermine production goals and profit margins. The ability to predict faults allows for critical parts to be replaced or maintained during off-peak hours or as part of a scheduled maintenance program. ADAPA® is used by factories and machinery manufacturers as the deployment vehicle for predictive models that constantly monitor equipment performance and alert anytime deterioration and brake down patterns are detected.