Interesting comments - as I remember the theory about structure, two
things stand out relative to the present debate:
1. Matrix organization and the project management approach are related,
but not the same. In matrix management the two-boss approach remains
always in place, although one of those bosses may change depending on
project assignment; whereas in the project management approach project
teams are developed as needed for those projects that are not well managed
within the conventional structure. One functional manager MAY be in
charge of the project managers/engineers, who in turn are part of teams
recruited for relevant functions. Not everyone with similar
responsibilities need be in project teams; the prevelance of teams will
depend on need and perceptions of upper management.
2. True matrix management is very expensive in terms of number of
supervisors and the need for special HRM-type structures and provisions to
overcome the inherent conflicts and discontinuities. Therefore, matrix
management should generally be used ONLY when it is the only way to manage
the great variety of complex projects the organization has to deal with.
Avoiding missed tasks and the costs (including lost reveneues) are the
primary justification for use of the matrix.
Control of resources IS key in any organization; most faculty inherently
know that, particularly those who are part of those institutions that are
or were managed in a completely top-down fashion.
For an example of the control issues in academia; when in industry I could
prepare appropriations requests that justified spending millions of
dollars, and my signature was accepted for requisitions involving $100,000
or more; yet now I'm questioned closely as to whether the $75 I put down
for transportation to and from an airport in another city is the right
number. And the whole travel request, out of a grant which could only be
spent for professional development of myself and my co-author, and which
required that I take that trip to present the results, was almost
disapproved because of that point.
Tim Edlund, Morgan State University
On Sat, 23 May 1998, fred nickols wrote: (in part) (my comment at end -
Tim)
> Most of the time, organizational structure is irrelevant; its real function
> is to balance power by way of balancing control over resources. (How's that
> for an outrageous statement?) Said a little differently, almost any form of
> structure can be made to work reasonably well -- if the relationships among
> people are solid enough to work through and around the "kinks" that any form
> of structure will introduce. Absent strong interpersonal relationships, no
> organizational structure can be made to work well.
>
> The raison d'etre for the matrix lies in its potential for maximizing the
> efficient use of resources. This potential most often and most clearly
> exists in an organization undertaking project work (e.g., big engineering and
> construction firms like Bechtel, Stanley, etc). By matrixing expertise
> against projects and separating direction from development, resources are
> both efficiently deployed and carefully developed. That's the theory. In
> practice, it ain't that easy.
The classic example was/is Boeing, with the various design and build
aircraft programs in various stages of completion. It's my understanding
that at least some of the big construction firms has stuck with the
project management approach, although the trend toward hiring such firms
as the owner's representative may have affected this more recently.