A Workflow Engine Server for the Design of Adaptive and Scalable Workflows

Abstract

Workflow generation is a widely used tool for business management. However, traditional solutions are primarily designed to generate a graphical representation of workflow elements rather than automate workflow generation. Additionally, current solutions require human input to generate a workflow that is designed to support a single task or job. These factors mean that current workflow generation tools are not flexible in terms of tasks supported, can't be easily scaled, and are not self-adaptive. This paper presents a proposal for a workflow engine server (WES) for workflow generation, which applies business workflow ideas to dataflow management. The WES architecture enables the automatic generation of workflows that are not restricted to a specific task or job. Initial evaluations of the proposed architecture demonstrate its potential for adaptive workflow generation, without the need for human intervention, for both small-scale and large-scale tasks.