One of the major challenges in implementing AI (ML algorithms) for project management is having solid knowledge, experience, and processes for leading project management activities. Project management offices (or project managers) cannot get the benefit of ML if they do not excel at the fundamental concepts, tools, and techniques related to it (e.g., planning and forecasting, risk assessment, tracking, monitoring, etc.) ML algorithms do not operate by themselves; they require training from meaningful data, an understanding of the business case and/or rules, and being aware that not all problems are solved by using ML. Saving Changes...
Depends on what the scope of the ML tools will be and the nature of the company's projects. If the company's projects are similar enough to those of others in their industry then an ML tool suite with sufficient external data might still be valuable even if internal data is inaccurate or incomplete.
If the company's projects are highly unique then it does raise the ante as far as having quality data which is certainly aided by a high level of organizational PM maturity but not guaranteed as poor quality data might also result from measuring the wrong things or PPM/EPM tool limitations.
Kiron Saving Changes...
BAGNIH Yves GBATIProgram Manager| AptivWest Bloomfield, Mi, United States
Very important question Antonio and very good response by Kiron. It is very contextual and it is a question for the planning phase of the ML initiative. Saving Changes...
I don't think the maturity level of the PM organization is nearly as important as the quality of data of whatever the thing is that is being managed by the PM.
...
1 reply by Antonio Collante
Sep 10, 2023 10:23 PM
Antonio Collante
...
Hi Keith,
How do you support your opinion? The question is implementing ML to support project management activities, rather than executing a ML project.
I don't think the maturity level of the PM organization is nearly as important as the quality of data of whatever the thing is that is being managed by the PM.
Hi Keith,
How do you support your opinion? The question is implementing ML to support project management activities, rather than executing a ML project. Saving Changes...
ML and PM both involve processing data to predict a path with the best probable outcome. Ultimately what we are trying to judge is the business performance. The better your dataset, the more ways you have to approach a problem and the better you can trust the validity of your analysis.
A company with virtually no PM maturity could implement ML based improvements if they have the data to analyze and the know how to do it. PM maturity won't help you if you lack the data to analyze, however. You need the data, before you can use the tools. Saving Changes...
