Precedence constraints in assembly line balancing
In one sentence
Precedence and resource constraints determine whether a task allocation whose times appear to add up can actually be implemented.
By the end of this lesson
- Identify a missing or duplicated task and a reversed precedence in an allocation of twelve tasks.
- Compare the usage intervals of a shared fixture at two stations and calculate the overlap.
- Classify tasks as accepted or rejected in terms of operator competencies and indivisibility conditions.
- Compare a second allocation, corrected by resequencing, with the first feasible allocation.
Why it matters
Precedence constraints state which tasks must be finished before a task can start. In line balancing, however, task sequence is only one of the acceptance conditions. In a list where the times fit, the required fixture may be in use at another station; the operator taking over a task may not have verified competency for it. If a part cannot be left half-finished, splitting a task on the drawing is also invalid. This lesson checks a prepared task allocation against such conditions. The aim is not to teach a new diagram or heuristic sequencing rule. It is to show clearly what evidence supports the accepted candidate and which condition is violated in the rejected one.
A predecessor and its successor can be at the same station; the predecessor must genuinely finish first. If the predecessor is at an earlier station, the arriving part must have that task completed. Work cannot start on the basis of a predecessor left at a later station. Remember also that different parts are being processed at different stations at the same time. While each part's task sequence is preserved, two stations may request the same shared fixture at the same moment. Precedence is therefore checked against part identity, and fixture use against the actual time interval. A list that looks sound for a single part can produce a resource conflict in the line's steady repetition. Preserving both views is necessary to understand whether the allocation is feasible.
A competency check is not about labeling a person fast or slow. Verifying that K01 can perform a particular task safely and at the required quality may be necessary before that task can be assigned to them. Training being planned does not mean the training is complete and the practice verified. Do not assume that a field left unclear in the constraint list is acceptable. The same principle applies to fixture access and task divisibility. A lack of information is not a lack of constraints. A candidate may be held pending missing information; this is not the same as declaring it definitely impossible. The evidence required for acceptance is stated in the decision record.
When verifying an allocation, no single percentage or highest workload can carry the whole decision. First it is shown that every task appears exactly once, then that sequence, time, equipment and competency conditions are met together. One violation is not offset by another good result. A candidate with a fixture conflict is not accepted because its columns are more even. If a correction is made, all checks are run again, because the task that moves may create a new precedence or access problem. Having two feasible candidates is valuable, but the decision rests on explicit conditions. The goal is not merely to fill the schedule but to perform the same work reliably at every repetition.
Formula
Station workload (s/unit) = Sum of assigned task times
Remaining time (s/unit) = Target cycle time − Station workload
Fixture overlap (s) = max(0, min(End 1, End 2) − max(Start 1, Start 2))
Assignment integrity = Every task must be assigned exactly once
Precedence condition = For every part, all predecessors must be completed before the successor
Unlock the rest of this lesson with an Academy account
Create a free Academy account with your work email address; it can be used once it is approved.
- Step-by-step instructions
- Worked example with real numbers
- Where the method misleads
- Common mistakes