Competing with the Future with Intelligent Agents…and a Confession
Creative Destruction – What Came Before What Came Next
Before the beginnings of Internet-based e-commerce, the computer was essentially a filing cabinet, handling back-office business, programmed under the metaphor of the clerk: capturing, storing and retrieving records. Today, however, we see not only Sun Microsystems’ first vision that the network is the computer, we can now see that the network is the business!
To participate in the digital economy, the business tasks we need to ask computers are shifting from back office record keeping to the front line business operations. We automate very different kinds of things: business processes and workflows which are knowledge-based human phenomena; and opening up our core business systems for direct interaction with customers, suppliers and sometimes with competitors (i.e. the airline industry combining flights to complete an itinerary). New forms of ubiquitous communications have enabled forward-thinking businessmen to blur the boundaries of industry and create virtual societies whose core competencies are information and knowledge management, taking a large part of the operational information once in the heads of the workers and codifying it in executable software.
New business realities have created new imperatives for enterprise information systems. Today’s enterprise systems must provide enterprise (and cross-enterprise) reach so that islands of disparate information can be integrated into a meaningful whole. They must be able to cope with the overwhelming complexity of distributed technology and a cross-enterprise information base. They must be open to survive in a network-centric ecosystem. Rapid application development goes without saying, and applications must be designed to adapt to constant change. Business systems must be knowledge-based (not just information-based) if they are to deal with the incompleteness and ambiguity of real-world processes and workflows. And they must be adaptive to meet the needs of the moment and bring productivity to an increasingly busy professional user, and self-service to our customers. This is what is being asked of IS today, a major challenge.
Business process management (BPM) technology and agent paradigms focus on solving cash and complexity. Intelligent agent technology is the next logical step to advancing the BPM technology paradigm and overcoming some of its shortcomings. Both technologies have been around for a while, but only now are they being repurposed for businesses in response to a rapidly changing world.
While the retail industry has been slow to grasp the importance of the Internet, until it Amazon, we are witnessing today a radical change in information technology. It’s called the Cloud, and cloud computing has unleashed a perfect storm in business. But unlike the retail industry which got Amazon, the Cloud will affect every industry and moving beyond enabling a virtual enterprise and moving to multi-enterprise virtual business networks and ecosystems. And it won’t just involve one cloud. Just as the Internet is a network of interoperating computer networks, the “Intercloud” will be an interoperating network of clouds: private, public and hybrid. Complexity will grow exponentially, and tools and methods are desperately needed to build enterprises that are complex, adaptive systems capable of bringing order out of chaos (chaordic systems).
Today, companies are wondering how to adapt software cash and flourish in complex Cross-cloud economy. Business leaders know all about rapid change and increasing complexity, and some are now asking “Why can’t software change to keep up with business changes?” It is this business case that is taking intelligent agent technology out of the research lab and military establishment and reorienting it towards competitive advantage.
Few computer scientists have not heard of software agents. Thanks to the dramatic growth of the Web, the computer science literature is full of discussions of software agents. Like tech terms, however, the term is tired of overuse and abuse. Various names have been used – knowbots, softbots, personal assistants, software agents, intelligent agents. And various secular definitions have been offered: software that thinks, software with a head, and an intelligent computer program.
Getting away from technology for a moment, the everyday term, agent, gives a starting definition: “one who acts for or in place of another”.A software agent is a software package that performs tasks for others, autonomously without being controlled by its master once the tasks have been delegated. “Others” can be human users, business processes, workflows, or applications. If you use a word processor such as Microsoft Word, you are a user of autonomous agent technology: spelling and grammar checkers perform these tasks for you, autonomously, as you type !
Is it a software agent or just another program? A basic software agent is based on three pillars, three essential properties: Autonomy, responsiveness and communication skills. The notion of autonomy means that an agent exercises exclusive control over its own actions and state. Reactivity means feeling or perceiving a change in one’s environment and reacting. And, even the most basic software agents have the ability to communicate with other entities: human users, other software agents, or objects.
Add to this definition the ability to plan and set goals, to maintain belief patterns (their own beliefs and those of other agents), to reason about the actions of itself and other agents (including humans) , and the ability to improve knowledge and performance through reinforcement learning, then you have the basic ingredients of a “intelligent agent”. An intelligent agent represents a distinct category of software that incorporates local knowledge about its own tasks and resources and those of other agents, allowing it to operate autonomously or as part of a community of cooperative problem solvers (including human users), each agent having its own the roles and responsibilities. But all these agents must be linked into a dynamic, meaningful and often changing whole. Say hello to goBPM.
the goal oriented BPM system (goBPMS) with choreography (not just old-fashioned process orchestration) will be at the heart of multi-agent systems that aim for the common goals set by goBPMS. An informative goal-oriented method is described in the column BPTrends, Goal-Oriented Organization Design (GOOD).
Goal-Driven BPMS with Agent Choreography
With careful planning and a solid understanding of the current limitations of intelligent agent technology, powerful enterprise information systems can be developed that leverage the Internet and distributed objects to gain strategic business advantage. Although the design and construction tasks are not trivial, information systems based on intelligent agent technology are essential. Learning how to harness this emerging technology should be a top priority for today’s business and technology leaders. Businesses can go it alone, but those facing immediate critical needs will be well advised to seek outside assistance from those that have gone before them.
Confession and some conclusions
First the confession. Parts of this column were written over 20 years ago by me and Faramarz Farhoodi (formerly at Logica, UK where he had extensive experience with the distributed intelligent agent system, CADDY) in the form of a two-part article for Distributed object calculation Magazine (I changed some words like BPR to BPM). What gall!
Well, thanks to Larry Roberts and his work with GTE Data Services, I had the opportunity to work with a packet-switched network that allowed computers to talk to each other in the early 1970s (no, that’s not was not the Internet, but the X. 25 allow computers to share packets of information)! Our whole team was full of excitement and wonder, as described in this To watch One-minute clip of futurist Arthur C. Clarke in 1974, long before the Internet and Cloud revolutions.
Whoops. Nothing seems to happen as soon as it’s “invented”, so we had to wait for Jeff Bezos to “Amazone” the retail industry before the Internet (invented 20 years earlier) has become the backbone of e-commerce. Amazon is a “virtual enterprise” marvel, leveraging intelligent agent technology, including its early use of Eyes and Ears, to provide services that only humans could previously provide. And, of course, in 2018, he delivered the richest man on earth at $157 billion in September 2018 (Forbes)! Oh, and Amazon just became the second company to have a market capitalization over $1 trillion! Hmm…remember Borders Bookstores?
Why am I bringing this material now? Well, as William Gibson once wrote in the Economist, “The future is already here – it’s just not distributed very evenly.” He is NOW it is time for intelligent agent technology, a technology that has been around and progressing for years, to take center stage in the business world in general, and specifically in the world of BPM, because business processes are how work was done in the age of the horse and buggy and how work is done in the emerging intercloud economy. Don’t just take my word for itnowof that. Here is an upcoming MIT 2019 event: Artificial intelligence is changing all businesses: don’t be left behind! For a complete new guide, feel free to refer to the book (written by Jim Sinur, James Odell and Yours Truly): Business Process Management: The Next Wave (Harnessing Complexity with Agent Technology).
Carpe diem!… or else!