At the end of the day, businesses need to be responsive to their customers so they keep coming back. With recent shifts from face-to-face to self-service business models, better, faster client experiences are not a luxury – they are a requirement. Automated decisions
are key to ensuring client experiences are consistent with what customers have previously known from face to face business while also allowing the business to adapt to market changes quickly.
With decision automation, business policy logic can be extracted from applications to enable business experts to implement decision policies, test, and execute changes without disrupting application code. This allows a much cleaner, more agile link between a policy change occurring and the implementation of that change as seen by customers, helping organizations be more responsive to changing markets and customer requirements. In addition, it also frees IT teams to focus on developing new capabilities for the customers that are revenue-generating rather than performing costly, ongoing maintenance. As a result, the overall business benefits from increased operational efficiency, transparency and compliance.
When automating decisions, it’s important to ensure that you are making the right decision tailored for each customer interaction. Traditional decision management translates business policies of the company into business rules. These rules, which often closely match the language used in the policy documentation, execute against the information provided by the customer and a decision is made based on these policies. However, if you think about your face to face interactions, there is typically far more at play in than simply the applying of the business policy. Often the experience and insights of the customer service representative (often learned over the course of their career from previous interactions) plays a key factor in the outcome of the decision. So how do we reproduce, or even improve on that, in the world of automated decisions?
AI and Machine Learning can help replicate that experience in an automated way. For example, if we pass into a machine learning system all the relevant past data we have on previous bank loan customers and whether they have defaulted on their loan or not the machine learning algorithm can search for patterns. In this way it can “learn” what factors are important to the likelihood of a customer defaulting, in the same way an experienced customer service representative would have built up a level of judgment to apply in these cases.
By combining the abilities of rule-based decision systems with the insights that can be gained from predictive machine learning, we can create automated business decisions that give both the best customer experience and create the best outcomes for the business.Intelligent decisions in the Z environment
If you have applications running in a Z environment that rely on high-speed processing – for example, banking, insurance, healthcare, or government transactions – the requirement for precise, automated decisions at speed is amplified.IBM Operational Decision Manager (ODM) for z/OS
and IBM Watson Machine Learning for z/OS
are tightly integrated with an optimized, shared high-speed interface between the two. This integration allows client applications running in CICS, IMS or z/OS batch to implement intelligent decisions with the superior performance you expect from Z.Learn more
Learn more about how ODM for z/OS and Watson Machine Learning for Z work together to help your organization make smarter decisions by attending “Make smarter decisions: Apply intelligence to your Z applications with digital decisioning
” on Tuesday, March 29 at 11am ET. To learn more about ODM for z/OS, join the decision management group
on the IBM Business Automation community