4.1 Project Management Discussion

Much of our preliminary work was spent in learning and implementing RUP, administrative planning of the project, and some very laborious manual analyses.

Spending the time to set up a well-organized project plan will allow the project to move forward more smoothly in the next phase. We evaluated our management of information flow, task assignment, and timesheet tracking. Initially the information flow was done through individual meetings and e-mail messages. This was not sufficient for information recall and organization.

Regular project team meetings started August 4 and were conducted weekly with minutes being posted to our FlowDx blog or sent out in e-mail messages. We moved to Google docs as a repository for our documents with version control for editing. In October, the focus shifted to presenting the current data to our collaborators for their approval, and the FlowDx.com website was organized as a place for the progress reports. Throughout November, we have worked exclusively on producing the project plan with all the RUP elements that apply to our project.

Reporting of tasks and time spent toward the FlowDx project has evolved through this stage of the project. We started with free-form reporting of the tasks and hours worked through e-mail. This was not efficient, so we next tried tasks list assignments for each team member, to be returned to the project manager with hours spend and status of the task. At first the task lists were exported from OmniPlan. These were not seen as helpful and were turned down as the supported method. Next was a hand-edited list of tasks sent to each team member via e-mail. This method was also not supported by the team. The next mandated method was when a team member turns in a deliverable, he or she also turns in a timesheet describing the task, deliverable, and the time spent.

One blatant lesson learned during the Elaboration Phase of the project was that our project plan and schedule omitted constructing a specialized data repository. Through our initial manual analysis of the use cases and automated classifiers, we saw an increase of the dimensions of our test matrices, the complexity of the test parameters increased, and the expected level of computation required, and therefore the need for a smart repository with integrated analysis tools. As we increase our emphasis on infrastructure, we are putting the use case analysis on hold. This iterative evaluation and readjustment is clearly the intent, and shows the benefit, of RUP. The design and implementation of this tool will provide support for higher throughput and more-meaningful studies and it has lead to a modular structure. The modularity of this tool will enable FlowDx, Inc. to bring the newest and best automated solutions to clients for their clinical flow cytometric assays. Many of these new ideas are being studied across academia, but there is no other company focused on bringing these solutions to the customer as a high-throughput solution to speed the analysis and minimize the manual analysis time.