Great post/thread on Mathematics, PM, and complexity

Glen Alleman and a number of commenters contributed to a great thread on math, PM, and complexity (here). 

I try to keep the ideas of complexity “science” in mind when planning strategy and its execution.  In particular, I have a deep respect for the power of self-organization and the need to create flexible rather than brittle management systems.

However, I’m not sure how powerful CAS really is as a theory, at least w/r/t/ project management.  For example, how do its predictions advance my estimation approach beyond what we’re doing w/ probability distributions (e.g., Monte Carlo simulations via Crystal Ball)?  To I really need math beyond that to get “good enough” estimates?

About these ads

3 Responses

  1. The challenge with PM and complex systems is based on the very nature of complex systems. Complex systems are deterministic in nature but behave in ways that cannot be predicted. However, once an end-state has been reached a backwards-looking assessment can show the chain of events that let to the end-point. It’s tempting to think that since the backwards look can see clearly that there could have been an efficient, forward looking plan to reach desired goals. (At this point those with a simplistic view of complex projects turn the project manager over to the Inquisitor for more lessons-learned sessions.) Complex systems also get confused with chaotic systems – but that’s another story.

    There basically is no way to apply scheduling and cost control theory to a complex project until an equilibrium has been reached. Said another way, complex projects are developmental in nature which means there is a degree of openness in the scope of work (connectedness between required features/deliverables NOT simply a wish list) and the risks involved. Experimentation is needed to get to some closure regarding scope and the risk terrain needs to be nailed down. If that can be achieved then classic schedule & cost control tools can be used. Monte Carlo and other methods depend upon this closure of scope,schedule, risk to be effective.

    Where complex adaptive behavior comes into play with projects can be seen in what army ants do – send out scouts, explore, and when something of value has been found the organization adapts by choosing the best structure (project plan/schedule/etc.) that optimizes the exploitation of the situation. At the point of deciding how to exploit project management comes into play and a linear plan can be developed.

  2. Hi Gary,
    Awesome comment…the last paragraph highlights the challenges of truly complex projects:
    1.) We can only generate linear plans for what are non-linear problems. As such, the plans are valid for shorter periods and must be refreshed more quickly.

    2.) We aren’t ants (too few and too expensive). Therefore, we need to simulate their “exploratory” behavior by engaging in shorter, wider-ranging (across and up/down organizations), and more frequent communications.

    3.) Even well-defined capabilities and scope can generate complexity if they’re large or new enough. The work required to deliver the scope consists of dependencies that are only discovered as the work is done or planned in more detail.

  3. Paul,
    I think this is why in IT scrum and agile are so popular. They accept two realities:
    1. complex systems are unpredictable strategically so teams work in a tactical manner, and;
    2. stakeholders frequently fall into that simplistic trap and generate unrealistic expectations as to the long haul and what things will look like.

    Construction seems to have realized this difficulty decades ago and are more rigorous in dealing with clients. Although, there is design-build in construction which has some of the spirit of scrum and agile – at least in the planning stages. They do accept limits since it’s pretty hard to move a footer once it’s been poured.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 2,518 other followers

%d bloggers like this: