Q&A: Why Business Process Management and Complex Event Processing Are Better Together

Business process management and complex event processing now support a broad range of business processes. How can these two technologies complement each other to further optimize business processes?

Business process management is a structured methodology for optimizing your business processes, while complex event processing takes action when a series or pattern of events is identified. Many organizations are now recognizing the combined power provided when the two approaches are integrated. For insight, we spoke with David Shaffer, vice president of Oracle Fusion Middleware.

Enterprise Strategies: What are BPM and CEP?

David Shaffer: Business process management (BPM) is an approach to thinking about your business and is much broader than just tools, and technologies. BPM enables a structured methodology for organizations to model, optimize, automate, and monitor their business processes to deliver higher operational efficiency, better business visibility, and greater agility. BPM products span modeling tools for business analysts, developer tools for system integration, business activity monitoring for dashboards, and user interaction for process participants. BPM has been around for a long time (in computing years, anyway), but some of the newest developments in BPM technologies are the emergence of transformational standards such as BPMN 2.0 and unified platforms to support processes that include applications, people, documents, and even trading partners.

Complex Event Processing (CEP) describes technology that enables organizations to detect patterns and trends in real-time across disparate event streams so they can respond more quickly and efficiently to emerging opportunities or threats. Using CEP, organizations can filter, correlate and process the many different events happening across an organization -- from analyzing financial transactions and markets and monitoring security breaches to tracking RFID tagged assets and improving logistics.

CEP is a new technology and many customers need to decide how it might apply to them. How are customers using CEP today?

We’ve seen an increase in the diversity of use cases for CEP technology in the past couple of years. Whether it’s streaming stock data for financial services, call data records in telecommunications or real-time vehicle-location data for transportation and logistics businesses, more and more companies in all industries are finding that they need to handle large volumes of data in real time.

Financial services and telecommunications organizations were early adopters of CEP and gained strategic value from it. Wall Street traders were among the first to benefit from CEP – using CEP to search large volumes of real-time data to look for certain types of trading activity and respond in milliseconds. Whether searching for favorable trading opportunities or finding potential fraudulent transactions in real-time, CEP has delivered considerable value to financial services.

It’s not just the financial markets that are benefiting from CEP. We are seeing more organizations using CEP in the supply chain/transportation/logistics space to correlate all of the different events that come through different supply chain networks and sources. For example, CEP can help to ensure one-day-delivery guarantees from order date. It is also commonly used for mobile application location-based events.

With respect to location events, we see a number of customers leveraging CEP in combination with Oracle Spatial to provide location and proximity analysis. One organization tracks emergency responders against spatial data for fire and disaster areas. They use CEP so they can track when people enter and leave the region to ensure their safety. It is also used to monitor that responders are only in a hazardous area for an appropriate amount of time -- analysis that is complicated by the fact that both the responders’ locations and the regions affected change dynamically.

What is the business challenge that BPM and CEP are collectively trying to solve? How are they connected?

Over the last few years, BPM and CEP have moved beyond their traditional isolated roles and many organizations are now looking at how the two technologies can complement each other to optimize business processes in the context of real-time events. Together, they enable intelligent business processes that react to changing business conditions and provide continuous visibility that enables the agile enterprise.

As an example, we see business processes used with complex event processing in fraud detection, enabling a business to identify a fraudulent transaction in real-time as opposed to the next day. With the combination of BPM and CEP, this detection can be done “in-line” during the business processes that manage orders and transactions, so that the fraudulent transaction never occurs versus detecting it and trying to roll it back or recover lost goods after the fact.

Why are BPM and CEP better together? How can BPM and CEP complement each other to help facilitate real-time decisions for the agile enterprise? While BPM is about improving efficiencies, CEP extends this to the real-time world where there are many events happening in a very short time period. By bringing BPM and CEP together, organizations are able to look at just their business processes and optimize processes in the light of real-time events. This enables organizations to manage and optimize their business more dynamically, making decisions based on customer demand or changing market conditions. As an example, the proliferation of real-time events based on social media data feeds or location-based service events means companies can have a great deal more knowledge of what is in the minds of their customers at a given moment. This allows for real revenue generation opportunities in terms of promotions or proactive responses to negative experiences, for example, to be built into business processes.

What type of company should be looking at using both BPM and CEP? Where is it inappropriate or unsuitable?

Most companies in just about any industry can, and should, take advantage of both BPM and CEP. As I already mentioned, telco and financial services are two obvious use cases and poster children for these technologies. We also see this technology combination being particularly useful to a number of industries where they have early-stage opportunities to differentiate themselves such as supply chain, manufacturing, and high-tech companies that have very high volumes of Web events.

However, although CEP is an alternative approach to analyzing historical data, this does not mean that historical data can’t be used effectively to optimize business decisions and business processes. For example, CEP is not intended as a replacement for data warehouses. Data warehousing is the right approach for organizations that want to do sophisticated analytics and historical analysis over long time periods and large quantities of data. CEP coupled with BPM is ideal for more real-time data -- when there is a requirement to filter, aggregate, and correlate events and respond in an event-driven fashion.

Oracle has other related technologies such as RTD (Real-Time Decisions). How is that related to, or different from, BPM and CEP?

BPM, CEP, and RTD can provide a holistic and complementary solution for intelligent processing. They work great together with BPM as the backbone process layer, RTD enabling real-time decisions in the processes with sophisticated, business-intelligence based analytics and CEP for real-time event streams.

Oracle Real-Time Decisions (RTD) is a highly scalable service-oriented decision management platform for decision optimization. It leverages real-time and historical data, business rules, predictive models, automation, and self-learning techniques to deliver decisions that adapt over time. Frequently, Oracle RTD is used in e-commerce or Web self-service sites or with inbound contact centers to optimize the customer experience as well as revenue generated for customer service or on-line sales.

Let’s take a credit card company as an example. In the context of fraud detection, BPM, CEP, and RTD can work jointly to flag certain transactions and make decisions about how to handle those flagged transactions. BPM would manage the overall transaction management process. CEP would monitor credit card transactions with specific patterns (e.g., multiple physical transactions from geographically disparate locations) and RTD would use closed-loop automated predictive analytics to optimize the fraud management workflow to improve fraud detection rates. In this case, RTD would automatically learn from each interaction by updating its predictive models in real time and those self-adjusting predictive models would be used to define and implement the best decision management logic for managing flagged transaction. In this way, the overall business processes associated with both fraudulent and non-fraudulent transactions would be automated and optimized.

What are some of the common mistakes companies make in deploying these solutions? Is there any best-practice advice for getting started?

It sounds contradictory, but we see customers both trying to apply these approaches too narrowly or also too broadly. Companies make the mistake of compartmentalizing BPM, making it focused solely on human workflow or on the flip side looking at events as if they were unrelated to the strategic business processes.

Although business processes are at the heart of operations, BPM has traditionally been viewed as a tool that can only be applied to large, complex, and strategic projects. However, modern BPM technologies can help organizations address both tactical and strategic business change. BPM can be used to solve both the urgent, pressing problems of the day and the strategic challenges or opportunities that redefine a company’s business.

Organizations need to look at their businesses holistically -- the events and processes across the entire business -- and then map those to a modern set of technologies that support them in their entirety. Additionally, organizations deploying BPM and CEP technologies need to remember that it’s not just about the technology. To be successful with BPM in particular, it’s imperative to get buy-in across the business for reviewing and optimizing the business processes and to include the end-users who are engaged in processes. What role does Oracle play in BPM and CEP?

Oracle provides a complete set of BPM and CEP technologies.

Oracle Business Process Management Suite 11g is a unified and complete business process management solution; it supports all types of processes with a unified process foundation, user-centric design, and social BPM capabilities. Oracle Business Process Management Suite 11g enables organizations to engage business and IT users more easily in the management of core business processes and to simplify the complete business process lifecycle.

Oracle Complex Event Processing 11g helps IT build and deploy high-performance, event-driven applications to improve the effectiveness of business operations. It can filter, correlate and process multiple event streams to detect patterns and trends in real-time, enabling organizations to capitalize on emerging opportunities or mitigate developing risks.