Sequencing & Scheduling Process

WHY IS MANUFACTURING (ASSEMBLY LINE) SEQUENCING / SCHEDULING COMPLEX

Manufacturing sequencing & scheduling is known to be a very complex problem. It is an “NP complete” type problem – that is, a problem where a perfectly optimal solution is impossible. The factors that cause the complexity are:

  • Problem size: The number of units to be sequenced is large.
  • From 2,000 to 30,000 units per month.
  • A large number of combinations of models and features are present.
  • The sequence has to satisfy many rules and constraints.
  • From 10 to over 200 constraints
  • Hard constraints: cannot be violated
  • Soft constraints: some are of higher priority than others
  • Constraints can change with changes in model years, model mix, features, line capacities and so on.
  • A short time window is available to generate the sequence

Often it is important to generate alternative sequences or ‘scenarios’, which ,can be evaluated to select the best sequence. Each scenario can include changes in both demand and constraints.

Changes can occur after the sequence is established. For example, the demand or mix of colors and features can change. This creates a need to re-sequence or change the previously established sequence. The complex sequencing process has to be repeated for the same time period. Where multiple lines are present, orders have to be balanced across the lines.

EFFECTIVE SEQUENCING PROCESS

An effective assembly line sequencing process must do all of the following:

  • Place several thousand units in sequence.
  • Consider of all relevant rules and constraints.
  • Meet the start after (earliest line on time) and start by (latest line on time) for each unit.

The sequencing process should be quick and accurate. First, this ensures timely release of the sequence. Second, it allows for generation of multiple scenarios. Finally, it simplifies re-sequencing to respond to changes.

A sequence that contains errors or does not consider all rules and constraints can result in many problems – some of them are:

  • Loss of throughput – for example if the number of color changes is excessive.
  • Increased costs due to more setups or changeovers.
  • Increased inventories if feeder line or vendor constraints are not considered.
  • Late deliveries if due dates are missed.