Will the Critical Path Concept Survive?

In my experience, most of project management only matters to project managers, and new tools don't necessarily change who needs what information. There are always exceptions, but the majority of the time people just want to know if you're done yet, and if not, when and how much it will cost. They have their own jobs to do and don't have the time or interest to peek behind the curtain.

Do project managers still use critical path? Some. I haven't needed it in a while; lately I've had more product-focus than project or used project approaches where critical path doesn't calculate (you can't calculate a path that doesn't exist, yet, and if there aren't any branches in the path calculating critical path becomes redundant). If I had need to use MS Project, I would likely also be paying attention to critical path. AI wouldn't change that need.

You wrote: "AI can process more data and provide faster analysis than a human project manager."

You're not technically wrong, but I don't typically do this by hand. If I'm using a tool, like MS Project, and keeping it updated, I usually only have to enter a little information and click a couple of buttons to see the critical path. I'm not tracking my projects in AI, and until I enter data and ask AI to do something, it's doing nothing. It's only faster if it has the data and somebody asks it to do something with the data.

A major shortfall of MS Project is it's reporting capabilities, at least for the desktop version. Add an AI to MS Project, to supplement it's existing capabilities, that allows non-PMs to ask for data and that would be interesting. At that point, project managers might be getting the critical path from the AI. Most other people probably wouldn't ask for it.

I think this statement might be overstating AI's current capabilities:

"AI tools assess all the factors in real-time and notify the project manager in advance of issues."

It might be more accurate to say that AI tools assess the information they have available to them and can raise both risks and opportunities, identifying potential impact, based on the information available to them. With enough information, an AI could possibly estimate how early a risk or opportunity might present itself, but unless you're talking about an Autonomous AI that contains all the data you need, somebody still has to enter the data and then ask the AI to do something with it.

Don't get me wrong. I'm not a luddite and I like new tools. It's interesting and fun, even, to imagine the possibilities. It just seems that, in a future where an Autonomous AI has the information it needs to give you the most likely date that a project will be finished, if it has been programmed to do so, it will run a PERT analysis and calculate the critical path then answer your question without telling you how it got the answer. Chances are that you'll only care how it got the answer if the answer was wrong or if you have to repeat the process on your own.

Posted: Feb 14, 2024 11:16 AM

Aaron makes very valid points; there is little left to add.

"This concept is in serious need of modernization." I disagree. This assertion is akin to suggesting that Pythagoras' theorem requires modernization due to its development over 2500 years ago. While critical path management does indeed need to adapt to modern tools and evolving environments, it requires no more than that.

Posted: Feb 16, 2024 5:03 AM

Thanks for this insightful article

Posted: Feb 19, 2024 2:33 AM

@Aaron Porter with all due respect: The power of AI is scary and can be intimidating to professionals whose jobs it has the potential to... change. The second to last paragraph in your response says everything. You simply haven't seen and/or used AI enough to see how very specific it can get. Each one of the 5 point in the article are spot on.
@Eduard Hernandez I agree, but it most like an issue of semantics... the concept is, in fact, solid. It is the method that can be modernized with AI.

Posted: Apr 11, 2024 10:31 AM

@Jason Farley, I admit that I don't work with the bleeding edge of AI tools; most project managers don't. While there are certainly people whose jobs have been negatively impacted by AI, some probably prematurely, we're talking about project managers and critical path. With the current state of AI and project management tools, using AI to calculate critical path is going to be more work than using the tool the way it was designed, depending upon the tool. There are a lot of work management tools with project management features where AI would probably make it easier, if not possible where it wasn't before, and there are a number of solid project management tools where critical path is just part of how the tool works; you don't have to ask for it.

Save the ad hominem fallacy, please. The majority of project managers using AI for their work are dealing with generative AI, which does nothing until you ask. The few that are working with autonomous AI still require a person to do something or provide at least a little information before the AI can do anything. This isn't to say that AI can't do amazing things. It can. I didn't say it couldn't; that wasn't the point of my response.

With regard to my second to last paragraph about risk, I've used more than one GenAI to help identify risks and potential mitigations. I've found that some provide more comprehensive responses than my team, some less so, and all are faster which is incredibly helpful when you want to kickstart the risk conversation. I still trust my team to decide which risks we can accept or ignore and which need to be addressed in some fashion, partly because I'm not going to put enough information about my company into a public AI for it to provide an informed response. I was on a webinar using AI in testing, earlier this week, and the speaker made a great point - don't put anything into AI that you wouldn't put on FaceBook.

Give me a private AI, and we'll have a different conversation, but with what most project managers have access to today, and are likely to in the near future, I stand by what I've said. But, that could be because I'm coming from the perspective of my company using AI to enhance our work and provide more value, not to replace people and cut costs. People that work at companies more interested in reducing headcount to cut costs have reason to fear, and the root cause is people, not AI.

Posted: Apr 11, 2024 3:33 PM