I have worked on a project before that faced some major scheduling complications. In my case, we noticed very late in the project that a manufacturing device we had used early in the project had been malfunctioning. Consequently, we had to go back to basically the beginning of the project and redo many of our tasks. In this case, overestimating the length of certain tasks would not have really helped us. With that being said, it is probably best practice to still include some buffer time in your scheduling. Using a combination of historical data and recent expert estimates would likely lead to best estimations. Has anyone found any other useful tools for schedule planning and remaining on a schedule? 

Accurate estimation requires a combination of historical data, expert judgment, and quantitative tools. In project management, accurately estimating activity durations is a cornerstone of planning and execution. It serves as the basis for resource allocation, cost management, and timeline adherence, directly influencing the project's success. 

Techniques for accurate estimation:

1) Expert judgement: Professional experience can always help in judging and estimating the risks, shortcomings, and failure in the project. People who have done such a project will have hands-on experience of the nitty gritties of it.

2) Similar estimation: This is the information from previous projects. Any major risk and their management which have happened in the past can be helpful in planning for the present project.

3) Calculative estimation: This uses mathematical calculations to predict durations based on specific variables. Example if we need 2 hours to assemble one unit, we need 20 hours to assemble 10 units. Once the parameters are identified, historical data is analyzed to establish mathematical relationships or formulas that describe how these parameters impact the project’s outcomes.

4) Three point estimation: Three-point estimation seeks to provide a more realistic and reliable estimate by taking into account the inherent uncertainty and variability in project tasks or activities. Here 3 approaches are followed; 

   Optimistic- The best case scenario. representing the minimum possible time, cost, or effort required to complete a task. It assumes that everything goes according to plan, without any unexpected delays or complications.

    Pessimistic: The worst case scenario representing the maximum possible time, cost, or effort required to complete a task.

     Most likely- This is the most realistic estimate based on the project manager’s knowledge and experience. It represents the expected time, cost, or effort required to complete a task under normal conditions, considering both positive and negative factors.

5) Bottom up Estimating: Bottom-up estimating involves breaking down the project scope into smaller, more manageable components and estimating the key elements required for each component. These estimates are then aggregated to determine the overall project’s duration, cost, and other needs. It helps in accuracy, transparency, and detail.
To mitigate the risk of task dependencies, choosing between fast tracking and crashing depends on the project's needs and available assets. Fast tracking is often the first approach when time is pressed, as it doesn’t typically require extra resources—just a reshuffle of activities to happen in parallel. However, it increases the risk of errors since multiple tasks are being handled at once. Project crashing, on the other hand, is resource-intensive, calling for extra hands or hours, but it can be more controlled and less risky in terms of task management. The domino effect in project planning is a reality that project managers must embrace to achieve successful project execution. By understanding the significance of predecessors and successors, project managers can effectively allocate resources, manage dependencies, and maintain the project’s momentum.
In almost every case the requirement is not well defined from the beginning. Often add-on tasks or requests become part of the actual pipeline. It is better to have buffer time in the schedule. 

Having slack in a project schedule is quite important to ensure that projects have built-in time for unexpected delays without derailing the entire schedule. I think that consulting the experts is the most important method to ensure that the project schedule is made accurately. Comparing to similar past project can be helpful, but not always so applicable depending on how similar the past project really was and if that project was done recently enough to be relevant. Consulting experts ensures that the project timeline is estimated accurately and enough slack is given to prepare for unexpected risks. 

As mentioned in the lecture, medical device development projects differ from other areas due to their inherent uncertainties—specifically, undefined timelines and unpredictable costs. Unlike traditional project management, where schedules can be estimated based on well-documented historical data, medical device projects often face regulatory delays, evolving compliance requirements, and unforeseen technical challenges. These factors make scheduling particularly complex.

While techniques like expert judgment, historical data comparison, and three-point estimation are valuable, they may be less effective in medical device projects due to these unique uncertainties. Instead, an adaptive scheduling approach may be more suitable. Incorporating rolling-wave planning—where detailed scheduling is done for the near term while broader estimates are used for later phases—can help address unpredictable factors.

Regarding task dependencies, while methods like fast-tracking and crashing can be effective in many industries, they may not always be feasible in medical device development. Given strict regulatory requirements, parallelizing tasks (fast-tracking) could introduce compliance risks, and adding resources (crashing) might not be viable if highly specialized expertise is required. Therefore, building sufficient buffer time into the schedule is critical.

The estimation methods you presented provide a fixed schedule of expected task durations. Another approach that should be considered is dynamic scheduling through real-time data integration. By considering ongoing performance data using project tracking tools, scheduling can be continuously updated and refined as the project progresses. This way, uncertainty in task durations can be mitigated by adapting to the actual work rather than sticking to a final plan that may not be valid anymore. Has anyone considered real-time performance data in project scheduling? Were there any considerable improvements?

Ideally teams should rely on both historical data and expertise when estimating tasks but in practice this may be difficult, especially when historical data is sparse or somehow biased by other factors that may not be applicable to the current project. The single most simple and effective way to reduce the risk of task dependencies is to build buffer into the schedule, as this is effective against any kind of delay since it simply gives you extra time that you can spend. Another way to reduce the risk of the critical path going off track is to mitigate against risks that could cause schedule delays on the critical path mainly by having redundancies in both people and resources that can serve as backups. 

Instead of only adding buffer time at the end, it can be more effective to distribute smaller buffers around high-risk or high-variability tasks, especially those on the critical path. Building on the idea of real-time data, I also think iterative schedule reviews can help teams catch issues earlier rather than reacting once delays compound. The example about the malfunctioning device really highlights that not all delays are predictable, so flexibility matters just as much as accuracy in estimates. In those cases, having contingency plans (like alternative suppliers or backup processes) can be just as valuable as time buffers. I also agree with the point about industries like medical devices, where fast-tracking isn’t always realistic, which makes proactive risk identification even more important upfront. Overall, the best schedules seem to balance structured planning with the ability to adapt quickly when reality doesn’t match the original assumptions.

Scheduling and the methods for estimating timelines are a crucial part of project planning. As mentioned in the original post, some tasks need to happen before other and expert judgement can be framed against historical data to set reasonable task windows. I believe it comes down a lot to research; a lot of the planning phase is learning how you can go about getting tasks completed. Thus, it is useful to compare procedure times with similar tasks in a literature review. Further allow slack time in the schedule, this way unexpected delays don't force tasks overtime. In some cases, tasks can be completed at the same time in the case they are independent. Such as coding and printing by delegating work early on. During project planning, you should ideally also be aware of the final deadline and space out work accordingly, leave the completion of the project to an earlier time, and give a period before the deadline for final minor fixes. Critical tasks should be started as early as possible in anticipation of delays. That said, what applications are most commonly best for updating the team on task progress? 

I personally think it is better to build buffer time into the schedule whenever possible. A project manager’s role is to create a realistic timeline and ensure the project is completed on time, and part of that is anticipating that issues will come up. Based on past experience and historical data, they should have a good estimate of how long tasks take, but no estimate is perfect. Including buffer time gives the team flexibility to handle unexpected delays without immediately impacting the entire project timeline.

That said, buffer alone is not always enough. If a delay affects the critical path, project managers may still need to use techniques like crashing or fast-tracking to recover lost time. I think the best approach is a combination, build buffer into the schedule upfront, but also be prepared to adjust resources or task sequencing if things start to fall behind.

Do you think too much buffer can actually hurt a project by making teams less efficient, or is it always worth having that extra cushion?