<!––>Are your planners stepping on your schedulers’ toes? In other words, are they guilty of planning in your scheduling horizon?

While there are many reasons for missed delivery dates, the failure to respect scheduling horizons is a common culprit.

I think we can all agree that planning and scheduling are fundamentally different. While the scheduling horizon stretches no more than a few days into the future, the planning horizon extends for weeks or even months ahead. Another difference lies in the level of detail. In the scheduling world, schedulers create detailed sequences on individual machines. On planet planning, sequences are irrelevant and capacity is planned in daily or weekly buckets.

The difference in the granularity of plans and schedules leads to an inescapable conclusion: There is simply is no way a planning solution or planner can use spare capacity in the scheduling horizon intelligently.

For example, consider a situation in which your scheduler only schedules 96% of the available capacity on a particular resource. What should your planner or planning solution do with that ‘spare’ capacity?

My answer, of course, is ‘nothing’, but let’s assume that the planner does try to assign this ‘spare’ capacity to an order. Will the planned operation fit the scheduler’s sequences for the day? There’s no way of knowing for sure, is there? And if the scheduler decides that the operation doesn’t fit and postpones it, how would that affect the planned delivery date?

Given that the planner or planning solution has no idea how to use the ‘spare’ capacity intelligently, the most sensible option is to ignore it. Decisions in the scheduling horizon should be left to those with insight into the appropriate level of detail.

So why then do planners and planning solutions regularly assume that ‘spare’ capacity in the scheduling horizon is up for grabs?

I’ve no idea unless, perhaps, they suspect that the schedules aren’t optimal in the first place. In which case the better approach is to fix the problem, rather than add to it.

If you agree that the ‘ouch’ of at least some missed delivery dates may be traced to heavy-footed planning incursions into the scheduling horizon, how should planning and scheduling work (or dance!) together?

Here are a few suggestions:

1. Define a scheduling horizon for each machine that extends to the end of the last scheduled operation on that machine. (This enables a  dynamic and machine-specific scheduling horizon that breathes with the schedule.)
2. Use the end of each scheduling horizon to mark the start of the period in which planners can make planning decisions.
3. Assume that the schedules make optimal use of available capacity. If this isn’t the case, the scheduling problem should be fixed rather than aggravated by inappropriate planning decisions.
4. Ensure that the planning solution or planner operates on the premise that there is no capacity available on a resource until the end of its scheduling horizon. Only the scheduler is allowed to assign capacity in the scheduling horizon.

This simple and straightforward approach clearly defines the responsibilities of  scheduling and planning, and avoids painful missteps that could hurt your delivery performance.

Do you agree that scheduling horizons should be respected? How have you tackled the challenge of integrating planning and scheduling?