Market players that can respond to the critical market events faster than their competitors will end up as winners in the emerging new economy. In order to win the next phase in the never-ending market race, they will also need to integrate their business processes with those of their suppliers and business partners. Additionally, the ability to quickly adjust processes to better respond to one′s customers will also become a decisive factor in the new economy. In this type of economic environment, the latency between the initial market event (any kind of significant disruption to the market status quo) and a response from the integrated process chain cannot take months or even weeks. In this article, we discuss the deficiencies of formal existing process management approaches and propose an agile process management approach based on a sense and respond loop. We present an event-driven process management infrastructure called SARI (sense and respond infrastructure), which can respond and adapt quickly to changes of a business environment. 数据挖掘研究院
The Agile Enterprise - An Introduction
Dot-com, e-business, e-commerce, outsourcing, virtual enterprises, globalized markets, "the flat world" - the last decade presented a plethora of new trends and strategies recommended to enterprises to keep in touch with their markets and stay prosperous. A fundamental change in how business is made is taking place, or has taken place already, in many firms in various places around the world. Not only by developments in the United States, Europe or Japan, but particularly also by the pressure of upcoming nations such as India or China, it turns out that strategies valuable in the 19th and 20th centuries are not necessarily helpful in the 21st century. 数据挖掘实验室
However, beyond all hypes and recommendations, the way from traditional to dynamic or virtual enterprises eventually leads to a general structure of agile enterprises. This is particularly true for enterprises in the technology market. However, what does agile mean on the company level? And what are the implications for IT and process - and eventually also knowledge management? Agility is usually seen as the capability of a system to react to unforeseen developments and to handle unanticipated change. This necessity to stay agile seems to be the only undoubted "bottom line insight" most authors share, observing the behavior of today′s business ecosystem.
Agility for Business Processes
Let′s focus now on traditional business process management systems that are using formal process models which to govern the execution of business processes. These systems act as a centralized control hub and invoke functions in enterprise resource planning (ERP) systems and business applications based on a formal plan given by the process model.
One major issue of this approach is that the process model assumes that it corresponds to a high level of predictability of the world in which a given enterprise operates. Where change occurs gradually and incrementally, for example, a sensible process management approach would stress efficiency in doing what the market will want and minimizing the cost and expenses of making and selling it. Such a business will articulate its mission and policies, which will rarely be modified because doing so would be disruptive - and disruption is the enemy of efficiency. 数据挖掘研究院
However, in unpredictable markets, in which customers themselves become unreliable predictors of their future needs, adaptiveness must take precedence over short-time efficiency considerations. When customer needs and their behavior becomes unpredictable, enterprises must move their center of attention to understand those needs to survive. Adaptive enterprises require a systematic ability to search out, capture and interpret clues about emerging and as yet unarticulated customer preferences and business opportunities.
A problem with existing business process models (such as BPEL4WS) is that they have no awareness of ongoing and external information that is not directly part of the business process execution. Business process models have weaknesses in incorporating changes and external signals of the business environment that can significantly influence or interrupt the process execution.
Traditional business process management systems are closed systems because a process model formally defines in advance (a priori) the course of execution and does not consider any other context information that is not related to process execution.
However, many business environments are dynamic and networked and therefore have to deal with permanent change and ongoing events of external sources such as customers and business partners. Sense and respond incorporate changes as well as internal and external signals into the control of business processes by processing information that is at hand (ex tempore) from the business environment. 数据挖掘研究院
Managing Business Processes with Sense-and-Respond Loops
Enterprises store data about their customers, suppliers, competitors that is generated during the execution of business processes. Process participants must have instant access to information relevant for the current business context. To be effective, decision support must take a broader view of the whole process of decision-making that is embedded in business processes. One of the key weaknesses of the current generation of workflow management systems and decision support systems is their lack of integration.
A sense-and-respond system is able to continuously receive, process and augment events from various source systems. In near real time, it transforms these events into performance indicators and intelligent business actions. It automatically discovers and analyzes business situations or exceptions and can create reactive and proactive responses such as: generating early warnings; preventing damage, loss or extensive cost; exploiting time-critical business opportunities; or adapting business systems with minimal latency.
An event-based sense-and-respond infrastructure (SARI) creates for enterprises a new level of awareness about the ongoing activities within business processes. The awareness is founded in the ongoing processing and evaluation of events with the context and key information of the current state of the business. SARI uses this context information of events to combine it with other available enterprise data sources (e.g., data warehouses) in order to:
- Minimize the latency between cause and effect of business decisions.
- Minimize the latency for discovering business situations and exceptions.
- Deliver up-to-date performance data.
- Trigger proactive action to exploit business opportunities or avoid loss or failure.
- Improve the accuracy of planning systems in order to calculate forecasts based on the most recent data.
- Continuously sense customer signals and systematically detect and anticipate changes in what the customer values.
- Respond to unanticipated requests in unpredictable business environments.
- Find meaning in a flood of random events.
In order to provide these capabilities, SARI manages sense and response loops that can be divided into five stages (see Figure 1).
Figure 1: Sense-and-Respond Loops
Sense: This stage describes the current state of the business environment. Events are continuously captured and transmitted to the SARI system where they are initially unified before the actual data processing starts. 数据挖掘研究院
Interpret: In this stage, the captured events are transformed into meaningful business information, such as key performance indicators, business situations and exceptions.
Analyze: The next step is the analysis of the generated business information in order to determine root causes for business situations and exceptions. The goal of the analysis is to find the best possibilities to improve the current situation of the enterprise. The analysis also allows prediction of the business performance and risks for making changes to the business environment. 数据挖掘研究院
Decide: In this stage, the best option for improving the current business situations and determining the most appropriate action for a response to the business environment is chosen. This step can be automated with rules or by involving humans.
Respond: Finally, the decision made must be executed in the business environment by communicating the decision as a command or suggestion (e.g., by email), or by directly adapting and reconfiguring business processes and IT systems.
During the processing in these five stages, business information is continuously generated and decisions are made for which responses follow. The response has an effect on the source systems (from which SARI originally received the events) and, consequently, also on the performance and the success of the company. Because the five stages are part of the sense and respond loop, the events of actions as part of a response are immediately captured and processed again. With each cycle of the sense-and-respond loop, the system improves the information quality and analytical results.
The processing of events within these cycles enables a continuous improvement of the business operations and business intelligence of the enterprise. Figure 2 illustrates this process. Each cycle of a sense-and-respond loop processes most recent data from the business environment and gains new knowledge about the current business situation. The final outcome of each cycle is a response to the business environment - the effects of which are captured by the next run of the loop. The goal of each loop cycle is to improve the accuracy of the assessment for ongoing business activities in order to improve and proactively control business processes. 数据挖掘研究院
Figure 2: Sense-and-Respond Loops in Concert 数据挖掘研究院
For example: A company offers transportation services for various goods. In order to better monitor, synchronize and optimize the processing flows, the company uses mobile devices (e.g., mobile bar code scanners) to store information for every article about the current location and the status of the current method of transportation. SARI continuously processes and analyzes this data and calculates indicators, which provide instant and concise interpretation of essential business information, such as the current transportation time of a shipment, the current transportation costs or the utilization of the transportation vehicle.
Additionally, SARI detects situations early that are relevant to the planning and coordination of the logistics (i.e., delays of a freight or loading the freight into a wrong container). In the case of such problem situations, SARI commences arrangements in order to deliver the goods in time, such as changing the transportation route (e.g., choosing a more direct transportation route to the customer) or changing the method of transportation (e.g., express transportation services). 数据挖掘研究院
System Architecture for Managing Sense-and-Respond Loops
Putting the pieces together, we propose a multilayer, service-oriented architecture as shown in Figure 3 targeting the requirements of agile process management and sense and respond loops. As mentioned earlier, service-oriented architecture (SOA) has gained popularity as new software engineering paradigm in the recent years. It arose from the necessity of creating components providing clearly defined small pieces of functionality that later on can be assembled into complex (usually distributed) applications. Using the SOA approach, SARI is a pool of services with an event service bus for a robust and flexible information transfer between the services. Each phase in the sense and respond loop introduced previously is supported by special sense and respond services that can flexibly interact with each other with an event service bus. 数据挖掘实验室
Figure 3: System Architecture
For data management, we distinguish different types of data: historical data, real-time data and metadata. The data warehouse provides a comprehensive view of historical enterprise data. The real-time data store is used to manage the most recent business information generated by the sense and respond services. Periodically, the data from the real-time data store is moved to the data warehouse (e.g., overnight). Information integration provides a single view of historical data and real-time data, thereby supporting the combined processing of historical and real-time data. All SARI metadata is stored in a separate metadata repository. Typical SARI meta data includes descriptions of event structures, the models for sense and respond loops, and the configuration of the system and sense and respond services.
SARI further includes a monitoring dashboard, which provides an interface for end users. It gives an overview of the current status of the event processing during the execution of sense and respond loops, delivers reports from data available in the data warehouse and real-time data store and offers search tools for discovering event patterns in historical events. 数据挖掘研究院
Traditional business process management approaches are unable to move quickly when changes occur. We propose a more agile approach based on sense and respond loops for managing business processes in order to accommodate faster changes in business environments. 数据挖掘研究院
