I'm speaking at the Step-Auto '07

I am invited by the SteP-In forum to present a paper on dynamic project scheduling and management, at the STEP - Auto'07 to be held on 17th Jan, '07 at Hyderabad. STeP-In forum is a non-profit organization based in Bangalore, India and organizes many international conferences on software testing including test automation, test project management etc, all around the year. During this year's Bangalore summit, Michael Bolton will speak on 'Rapid software testing'. Please visit the forum's website for more details.

This is my second time at the summit, presenting a paper. The first time, I presented a paper on automated generation of test cases, at the STeP-In Summit '05 held at Bangalore. Details of my paper are available at my website here.

For those who are interested in knowing what my this year's paper is about, here is an abstract...

Project scheduling and tracking is no doubt one of the most complex processes of SDLC. Though, this is typically performed by the project manager, the direct impact of this is often on all team members. Project managers need to collect essential information to assess the health of the project all the while, which typically all lies with the team members. However, most times the team member doesn't have all the necessary data himself, but is spread out in the team. Hence, the project manager is often forced to interrogate and extract information from the team members. If not practiced with care, it can be a painful experience for both sides. Most of us would have gone through such, often 'bitter', experiences at work. Trying to work out a solution for this, I realized that, most of such problems arise from the inaccuracy or even inability of the team members to prioritize tasks, plan/schedule them and predict measurable impacts of those prioritizations on yet other tasks. Team members need to be more capable of predicting their own performance and plan themselves accordingly. In essence, empowering of the team members rather than centralizing the control with the project manager is the philosophy that is fundamental to my solution.

DSM (the tool) has its goal at helping team members, who report to any level of management, to self-manage their activities in a typical project environment. This said, the activities don’t need to be project related activities alone, but any including those unplanned, task overheads, vacation plans etc. Original intent of the tool was to track moving targets, which is nothing but calculating the remaining days of each of the activities based on the input status (% completion) and helping the user to derive the start and end dates for the remaining period for each of them. In addition to this, the tool supports dynamic scheduling and maintenance of these activities. To do this, it utilizes the very simple concept of 'current task'. Thus, at any point in time, only one of the tasks can be current, which also implies that the start date for the current task will always be today. All subsequent activities are processed in a sequential manner. However, the user has the option to make any other among the activities as current and process accordingly. Another possibility with the tool is creation of burn-down charts, which makes it in line with the agile way of tracking schedules. The tool also has certain ways to manage expectations on deliveries to customer, primarily by way of generating a weekly delivery report.

Apart from the above features, the tool also has a few features that facilitate typical project management. It provides metrics like comparisons of the planned and actual, in terms of estimates, status updates, remaining days and end dates of various tasks. This is facilitated through a feature that allows 'snap-shot'ing essential details of all tasks at any point in time. There is also the facility to split up overhead among planned/non-planned tasks and to generate neat useful charts out of it.