Managing Weather Stoppages in Schedules
“How do I account for weather stoppages in my project schedule?”
This question comes up a lot. Particularly from folks who work in a business where weather conditions can reap havoc upon your beautifully crafted schedule. The answer to mitigating weather stoppages can be a choice among several approaches. Which one you choose may depend on company guidelines, contract language or other variables. The trick is to figure out which of the following approaches is going to be the most acceptable. Acceptable to your customer, your boss or whoever else is calling for this contingency. And these are only a few options among many out there in use today.
The general principle is to mitigate the risk of weather stoppages in your schedule by building in some kind of buffer. This buffer will help you absorb the impact of weather-borne delays. Your first task therefore, is to quantify and document this weather stoppage risk.
To determine the risk level you are trying to mitigate, use the information from the National Oceanic and Atmospheric Administration (NOAA). They have schedules of monthly anticipated adverse weather delays for the area in which the project is being executed. These schedules map out the number of days per month that will be adversely affected by severe weather events.
Here is an example of an adverse weather table you might use to build a severe weather strategy.
Strategies for Mitigating Weather Stoppages Risk in your Schedule
Once you have quantified your level of risk, you need to build this into the schedule. There are several options for this, some of which are better than others. In order of least preferred to most preferred: here are some basic options:
Calendar Weekends
In this scenario we use weekend calendar nonwork days as a weather stoppages mitigation strategy. In other words, if we get stopped by snow or rain for a day or two during the planned working week, we’ll just work the equivalent number of days during weekends to catch up. This is common and among the most popular options because it’s simple and requires pretty much no special treatment of the planning schedule. You simply say it is so, and manage the delays as part of the status process.
It’s also the least effective as a mitigation strategy because it really doesn’t push anything out in the planning schedule. In other words, the schedule doesn’t reflect anything different and ignores the fact that, if there is a significant risk of adverse weather that month, say 6 days, you could easily go from having 20 working days to just 14 if the weather events occur during planned work days. In other words, you have no visibility of the impact of that 6 day risk should it be realized. Instead you have just declared weekends as your go-to buffer if the worst happens.
Weather Stoppage Activities
So if the Calendar Weekends strategy is not going to cut it, how about something that is more visible, more tangible. This is achieved by placing a “Weather Stoppages” activity on the critical path just before the substantial Completion activity. Its duration is calculated by adding together all the predicted adverse weather days for the period of performance.
This strategy effectively increases the length of the critical path to push out the project complete or Contract Complete Date (CCD). This is a better strategy than our first option because at least the schedule plan is truly reflecting the potential impact of these weather delays. It’s also easy to identify in the schedule. Also, unrealized weather days could be subtracted from the Weather Stoppage activity’s remaining duration as part of the status process. This would help keep a realistic CCD being reported for the project.
However, this method is slightly problematic if you have baseline change control protocols (which you should always have) because it could be interpreted as a baseline change depending on the capabilities of your software system. With Primavera you would just be careful to change the Remaining Duration only so that it is clear you are altering the forecast values, not the baseline values. It’s also a slightly higher maintenance approach for the scheduler.
Buffering Activity Durations
This technique would require the increase of activity durations by the number of weather days in each period of the project. Of course you would only do this with activities that can be affected by adverse weather and it would also reflect in the critical path and CCD for the project. However, this approach is less than ideal.
For a start the risk is hidden and cannot be easily pointed to in the schedule: hence it would require some backup documentation to explain where the buffering occurred, why it was done and on what it is based. While we should do this anyway as part of our risk management process, being able to point to a specific item in the schedule is also helpful.
The other problem I have with this strategy is the potential for slippage due to weather or other events that could potentially push higher buffers downstream into periods of lower risk, thus skewing the remaining work scheduling in an unnecessarily negative way. It’s also a lot of work to setup in major schedules with thousands of activities.
Calendar Nonwork Days Option
Last but not least is the option to use a so called “Weather Calendar”. In your schedule you assign this Weather Calendar to any activity that can be affected by weather stoppages. Scheduling Tools such as Primavera P6 and Microsoft Project both allow you to assign different calendars to each activity in the schedule. Therefore we create a copy of the default calendar and add weather non-work exception days to that calendar. This is typically done by making the last working days for each month of the Calendar Nonwork exception days. Here’s an example of that in Primavera P6 and then Microsoft Project.
The advantage of employing a separate “Weather Calendar” is that it can be applied to only activities likely to be affected by adverse weather, and it automates the planning and maintenance process in as much as activities using this calendar will not show working duration in these periods. They will effectively ‘step over’ the nonworking periods and their finish dates will be calculated accordingly. You quickly get to see activities being pushed out further into the future as a result of this mitigation strategy.
We can see a visual of this affect if we use the ‘Bar Necking’ feature in Primavera P6 based upon calendar non-work periods.
The weather nonwork days are ‘burned in’ to the calendar so to speak, and will remain steady within their periods, even if weather and other delays start pushing work out into the future. The other benefit is that with Microsoft Project and Primavera P6, actual dates can be recorded on non-work days. This means that if it doesn’t rain cats and dogs at the end of every month, as we planned it, we can still go ahead and enter actual dates that will pull in future work. In short, if all work went to plan, and it didn’t rain all year, we would automatically complete the project early be exactly the number of weather days we had planned on.
Summary
Whichever strategy you choose, or indeed come up with for your own situation, you should always do something to mitigate the very real risk of weather stopages on projects that can be affected by such things. Remember Mr. Murphy is always looking for a way in to your schedule, so you need plenty of Murphy Repellent sprayed around.
Sometimes, weather days will be a contractual requirement set out by your customer. Sometimes it will appear unrealistic when you build that contingency into the schedule; pushing dates out further than you would like and causing you to have to work the schedule in order to meet contractual milestone dates.
However, if it does rain, you’re covered every which way from Sunday if you got your mitigation strategy correct. So as always, you plan for the worst, while hoping for the best. Or as Eisenhower famously stated in an address to the National Defense Executive Reserve Conference in Washington DC, 1957, “Plans are worthless, but planning is everything”.
For the present though, I hope this article has given you some ideas for better planning going forward.