(photo by Joel Sadofsky)
One of the problems with developing AI capability for project management is the diversity of the types of projects. For example, there are construction projects, software deployment projects, and projects to change business processes. While the overriding principles of project management are the same, projects in different functional areas offer a unique challenge for AI. Machine learning, one of the components of AI, requires a significant amount of data to ‘train’ the AI tool. This can be in the form of historical data such as lessons learned. However, the lessons learned from a completed construction project may not be valuable for a project with the goal of deploying software. While some commonality may exist, it is more likely that an appropriate dataset from each project area is required.
The amazing value of project management knowledge such as the PMBOK® guide is understanding a generic set of principles that can be applied across any project. Although the principles are an essential element, AI requires specific details that allow an algorithm to consider relevant data in order to predict results and ultimately make an optimal decision for the project.
Given this situation, the availability of good project data might be as valuable as the ability to create an algorithm that can take advantage of the data. Is your organization collecting and storing project data?