Without attempting to generate a specific sequence for these tasks, we note that they seem to belong to seven categories of work.

Develop and prioritize requirements.

Design, build, and test hardware.

Design, write, and test software.

Set up production and service$/$repair systems with spares and manuals.

Prepare and implement a make$-$or$-$buy analysis.

Develop release plan.

Design marketing effort, with demonstrations, brochures, and manuals.

Based on this analysis, it would appear that the project will need the following elements:

Groups to design the hardware and software.

Groups to test the hardware and write and test the software.

A group to engineer the production system for the hardware.

A group to design the marketing program.

A group to prepare all appropriate documents and manuals.

And, lest we forget, a group to administer all the above groups.

These subsystems represent at least three major divisions and perhaps a half$-$dozen departments in the parent organization. The groups designing the hardware and the multiple operating systems will have to work closely together. The te groups may work quite independently of the hardware and software designers, but results improve when they cooperate.

Trinatronics has people capable of carrying out the project. The design of the hardware and operating systems is possible in the current state of the art, but to design such systems at a cost $010\%$ below potential competitors will require an advance in the state of the art. The project is estimated to take between $18$ and $24$ months, and to be the most expensive project yet undertal by Trinatronics.

Based on the sketchy information above, it seems clear that a functional project organization would not be appropriate. Too much interaction between major divisions is required to make a single function into a comfortable organizational home for everyone. Either a standalone project or matrix structure is feasible, and given the choice, it seems sensible to choose the simpler standalone project organization if the cost of additional personnel is not too high. Note that if the project had required only part$-$time participation by the highly qualified scientific professionals, the matrix organization might have been preferable. Also, a matrix structure would probably have been chosen if this project were only one of several such projects drawing on common staff base.

Questions

Consider the applicability of a "weak" matrix structure for this project. What would be the pros and cons of this approach?

Consider the applicability of a "strong" matrix or "balanced" structure. What would be the pros and cons here?