Discovery Machine KnowledgeNET is a solution for automating the knowledge work of many key individuals within the enterprise. A KnowledgeNET is a network of knowledge service gears that can dynamically call upon each other to carry out their activities. Our solution builds upon and extends the capabilities of SOAs in two ways: by deploying best practices of key individuals, and by providing services that “understand” their own inner workings.
SOAs are only as good as the services they provide. As they currently exist, SOAs are little more than a delivery mechanism for small applications (more accurately the functions that now comprise applications). In order for an application to be turned into a service, it must provide the SOA with information about how it is to be used. In SOA deployments nearly all of the services are programmed by software developers. This limits the return on investment that SOA can provide. The Discovery Machine KnowledgeNET, on the other hand, provides a knowledge-based SOA that enables key individual expertise to be leveraged throughout the enterprise. Knowledge service gears are executable services that carry out the best practices of key individuals.
Key individuals do not work in a vacuum but rely on the input of others to carry out their work. In the same respect a knowledge service model can draw upon outside resources to complete its tasks by requesting information from others through emails and online forms. The automation of knowledge work is accomplished by drawing upon existing data and information found within the organization.
In manufacturing, there is the assembly line. Automation on the line involves robotics at each step and there are different machines doing different things. Knowledge work, however, is not linear. It is a network of activities that are occurring in sequence and concurrently. These activities have been represented in swim lane diagrams and other forms. The business process modeling languages come in many flavors (BPML, BPEL), but the basic premise is to represent the workflow in a knowledge work setting. Unfortunately, these solutions are broad and shallow and do not have the depth of representation to automate knowledge work.
The cognitive representation used in knowledge service models gives the models a far greater access to what it is doing and why. Except for the top task, each task within the model is being done in service of some higher task by employing a method. The model is able to access itself, giving it a greater ability to know where it can be effectively used. The model understands its own inner workings, which means it can be carried out more effectively within the enterprise.
