Critical path scheduling is the act of applying a logical sequence (by defining constraints) to the activities
defined in the work orders. Most project management software employs a PDM (Precedence Diagramming Method) interface for defining
the logic network. The sequence of activities which have no float or slack (Float = 0 hours) is called the critical
path. It determines the remaining duration of the turnaround.
The first step to turnaround scheduling is to define all 'hard' constraints. These are constraints that must be honored.
For example, you cannot inspect the interior of a vessel until the manways have been opened. eTaskMaker® project planning software automatically generates hard constraint logic for you.
ATC Professional project management software automatically generates 80-90% of this logic
for you as well when creating the initial schedule.
It is not necessary (although it is not detrimental) to add redundant constraints such as:
- A --> B
- B --> C
- A --> C (this is redundant and unnecessary)
Activities can have multiple predecessors and/or successors. Activities can be started as soon as all of their
predecessors are completed. For instance, "COOL DOWN / GAS FREE" can have as successors "INSTALL TEMPORARY LIGHTING" and "INSTALL
ENTRY LADDER". Also, "CLOSE MANWAYS" can have as predecessors "REMOVE ENTRY LADDER", "REMOVE TEMPORARY LIGHTING" and "REMOVE
AIR MOVERS". Remember:
- Predecessors - the activities that must be completed before the next one can start
- Successors - all activities that follow a specific task.
Activities can start as early as desired, or can be delayed until they run out of float or slack, thus becoming
critical. At that point they are identified as the critical path. Any delay of the critical
path activities will cause an equal delay for the entire schedule.
Most activities will have float or slack, which is the amount of time they can be delayed until they become
critical (Float = 0 hours) and impact the unit's start-up date.
Realistically, activities that have very little float or slack should be treated as critical simply because
there may be a degree of error in the estimates. A sequence of activities with float = 5 hours could easily be critical if
their combined durations were underestimated by five hours (or the critical path was similarly overestimated).
Priority
Be sure to schedule all equipment inspections early. This is very important, because some findings could require
major repair work that might impact the schedule. All high manhour work orders should be started as soon as possible.
Some equipment will merit a lowered priority, if the past experience indicates little or no repair work will
be required. Consult the inspection reports to identify the extent of the repairs during past turnarounds.
Low priority work is usually classified as "fill-in" work. It usually includes all kind of small jobs - mainly
piping and valve work. You can spread out this work over the duration of the turnaround, to help smooth out the manpower requirements.
The scope of these small jobs seldom grows into a larger one, and has no probability of showing up as the critical path.
They may, however, in the aggregation of several jobs, result in a critical mass of work (that can not be finished
with available resources within the current critical path timeframe) and therefore eventually cause a delay in the schedule
(overtaking the critical path). Critical mass develops when the rate of progress is insufficient to complete the work before
the critical path end date. It is usually due to insufficient manpower. This is the reason for keeping a close watch on the
actual number of workers, every shift, and comparing it with the schedule requirements.
Sequencing the Work
After the basic schedule has been created, and the work prioritized (sequenced) according to an Operations /
Production equipment availability schedule and the other considerations discussed earlier, you should sequence the work in
such a manner as to enhance the utilization of manpower, tools and equipment.
In sequencing the work, we have to consider the type of job, the resources or skills involved and the physical
layout of the unit or plant.
The first step is to determine the number of crews. We do this by reviewing a resource histogram (utilization)
report for all resources and record the peak leveled number of craftsmen. So, we divide
by two to arrive at the peak leveled number of crews, and add ten or twenty percent. This is a good rule of thumb for preliminary
manpower planning. The reason you need to hire more men than scheduled is to compensate for absenteeism, dismissals, and additional
work arising from inspection.
You may have several crews of any particular resource; even if you only have one generalized resource/skill
designation such as "multicraft".
Start by sequencing the "hard" crafts that perform most of the mechanical work. These are usually Boilermakers,
Pipefitters, Welders and Mechanics. If you sequence these crafts properly, all support crafts will follow accordingly and
may not need to be sequenced.
Activities that are critical or near critical (having little float) should not be delayed, as the manpower required
to accomplish them must be supplied as dictated by the schedule.
We can sequence the work that has float or slack by tying or restraining activities together, in such a fashion
as to cause a crew to go from one job to the next as soon as the first one is completed.
The best way to this is with the help of a plot plan or equipment layout drawing of the unit / area. When sequencing
the work, try to keep the movement or travel between jobs to a minimum. Causing workers to continually move from one end of
the unit to the other is inefficient and can result in a significant waste of manpower.
Every time you tie or restrain activities to sequence manpower, check to see if that action resulted in making
the activities critical (or near critical). Near critical activities have very little float or slack. If the activities have
become critical, then it is best to undo the tie or restraint, otherwise you may be scheduling too tightly - increasing the
probability for an overrun.
This is a trial-and-error method, but it is not too difficult to achieve, and the result will be a workable
schedule with a realistic manpower utilization.
Efficient Manpower Utilization
Effective manpower use is achieved by eliminating:
- Wait time
- Movement (travel time)
The best way to achieve high efficiency is to sequence the work as described above, and then issue Shift Schedules
that list fifteen (15%) percent or more work than can be accomplished. This keeps the schedule sufficiently flexible to accommodate
the changing conditions that cause some work to not be available as scheduled
(lack of permits, lack of equipment, etc.). Field supervisors will then always have sufficient work scheduled to keep
everyone busy at all times.