Activities (except project complete) that have no successor are referred to as dangling activities. These open ended activities are loose ends in the project that make monitoring their progress difficult. There are, however, more than one way to tie up dangling activities in the schedule.
Relationships define the order and interaction between activities. The most common is the finish-to-start (FS) relationship where the successor cannot begin until the predecessor completes. A typical FS application is the relationship between ‘pour foundation’ and ‘install walls’. The FS relationship connects the predecessor ‘pour foundation’ with the successor ‘install wall’. Not only does this relationship tell us the order of activities, but it also stipulates that ‘pour foundation’ must be completely done before commencement of the successor ‘install walls’.
Sometimes though activities have no apparent logical successor. What do you do in these situations? Do you leave the activity hanging with no successor? The answer is no you don’t. Consider instead the alternatives.
This article explains alternative approaches to handling dangling activities in the schedule. Schedules presented in this article were generated in Oracle’s Primavera P6 Professional.
We have in Figure 1 our demonstration project.
Note, in particular, the ‘install fence’ activity. This activity has no defined successor. It is left dangling on its own with nothing to connect it back in with the schedule. Note also its 31-days of total float; install fence some may consider a high float activity. Because ‘install fence’ has no successor its total float is measured from its planned finish date to the end of the project.
This means in our example that install fence can delay 31-days without affecting the end date of the project. High float activities, in particular, are discussed in the blog The DCMA 14-Point Assessment and High Float Tasks.
While the defense contract management agency (DCMA) guidelines at the above link says high float activities have total float greater than 44-days many would consider 31-days total float to be a high float situation. Well, not only is ‘install fence’ high float, but, again, it also is a loose end activity.
Why might this be a problem? The dangling ‘install fence’ activity provides no early warning that its lack of progress is becoming an issue. You will not know there is a problem until all 31-days of total float are spent and it begins to push against the end date of the project. When this happens it’s too late. You’ll have a project fire on your hands, and you will have to scramble to complete ‘install fence’ to minimize any further project delays it is causing. But this project fire and delay is completely unnecessary. You had thirty one days you let go by without making any progress on install fence. In light of this, how can we tie install fence back into the schedule to prevent a project fire?
First, scheduling guidelines, in particular, the DCMA would consider install fence’s lack of a successor “missing logic”. Refer to the blog The 14-Point Assessment and Schedule ‘Missing Logic’ Inspection for a discussion on “missing logic”.
So install fence needs a successor to adhere to accepted scheduling guidelines. In an effort to meet guidelines the first response of schedulers often is to connect the dangling activity to the ‘project complete’ milestone, the end of the project. This satisfies the ‘missing logic’ guidelines, but still provides no early warning that your install fence delays are threatening the completion date of the project.
You really want the subcontractor to complete ‘install fence’ task much sooner than the close of the project. Well, then why not place a finish on or before (FOOB) constraint on install fence? Constraints are a way to place additional restrictions on a task date to emphasize the importance of that date in the life of the project. In Figure 2 we place a FOOB constraint on ‘install fence’.
Note that the total float decreases to 2-days when we apply this constraint to install fence, Figure 3.
Great! So install fence is scheduled to complete much earlier than the end date of the project. This is what we want. But there are drawbacks to using constraints.
Constraints should be used sparingly in schedules. One reason is that they do not automatically update with changes in progress. So a schedule with many constraints requires much more tedious attention. Constraints should also come with a note explaining their purpose.
One former student wanted to hide total float from his subcontractors because they tended to delay work until the last possible moment. Constraints can hide total float, but it is not an honest scheduling practice. Constraints should not be used simply to hide total float from your subcontractors. Is there then a better solution?
It is better to think through your schedule situation, and try to find a logical common sense successor to the task in question. In our schedule it would make sense to tie in ‘install fence’ to the ‘substantial completion’ task. The fence installation is one indicator that the project is near completion. In Figure 4 we connect ‘install fence’ to ‘substantial completion’ in a FS relationship.
This works well to eliminate ‘missing logic’ and to avoid a constraint. But our total float (21-days) is still in what many would consider a high float situation. Also, for the purpose of discussion what then can be done to reduce our 21-day float activity?
Let’s further examine the situation. Consider the install fence activity. What is the purpose of the fence? The fence provides security to the job site. When might we want security provided to the job site? Before the arrival of any high voltage hazardous equipment. In light of the fences security function the logical successor of install fence would be ‘install bus and jumpers’, which are hazardous electrical components. In Figure 5 we make ‘install bus and jumpers’ the successor to install fence and our total float compresses to 13-days, which is much more reasonable.
Scheduling guidelines will flag any dangling activity, i.e. any task (except ‘project close’) with no successor, for ‘missing logic’. Minimum strict adherence to this ‘missing logic’ guideline usually results in connecting the loose end activity to the ‘project close’ milestone. This meets the ‘missing logic’ guidelines, but still fails the high float guidelines.
Schedulers may resort to external constraints to reduce total float. This works well to minimize total float, but may be less than honest. Full disclosure is a characteristic of quality schedules. The better solution is to consider possible successor candidates. This way you will decrease total float, provide a logical successor, and maintain a dynamic and responsive schedule. So considering the purpose or function of a deliverable in question just might help to find a logical successor, and keep total float at a minimum.