Scheduling overview

Scheduling answers one question for every job: given the work in its routing and the capacity of the resources that do it, when does each operation start and finish? Epicor’s scheduling engine works this out by converting each operation into blocks of time and placing them on resource calendars.

What the engine uses

Input	Where it’s defined
Operation times: setup, production standard, quantity	Job method (copied from the part method in Engineering Workbench)
Queue and move time	Resource group or resource
Which resource does the work	The operation’s scheduling resources: a resource group, a resource or a capability
Available hours	Production calendars and shifts on the site, resource group and resource
Operation relationships and send-ahead	The job’s operations (Finish-to-Start by default)
Dates	The job’s Required By date (backward) or Start date (forward)
Priority	Job priority and the global scheduling order

Direction: backward or forward

• Backward scheduling starts from the Required By date and works back to find the latest start that still meets it. This is the “just in time” approach: material and labor are committed as late as possible, keeping WIP and inventory low.
• Forward scheduling starts from a start date (usually today) and works forward to find the earliest finish. Use it to answer “when can we have it?” or when a backward schedule would start in the past.

Capacity: finite, infinite and rough cut

Mode	Behavior	Good for
Infinite	Places load on resources without checking whether they’re already busy	Seeing total demand on each resource; fast what-if dates
Finite	Only uses capacity that’s free; later jobs move out to where capacity exists	A realistic, executable schedule for the near term
Rough cut	For jobs due beyond the site’s Rough Cut Horizon, schedules infinitely using lead times and doesn’t record any load against resources	Long-range planning without slowing the scheduling run

Resource groups and resources have a Finite Capacity flag and a Finite Horizon (in days from today). Only work inside the horizon is scheduled finitely; beyond it, the engine switches to infinite capacity, so a job due next year doesn’t compete for capacity today.

Rough cut in practice

Rough cut only applies to jobs whose Required By date is beyond today plus the horizon (set on the site’s planning settings). Jobs inside the horizon get a normal schedule with full load. If you test rough cut with jobs that are all due soon, you won’t see it take effect.

Global scheduling

Scheduling one job at a time in Job Entry is fine for small changes, but it schedules jobs in the order you touch them. Global scheduling reschedules everything in priority order:

1. Calculate Global Scheduling Order selects the candidate jobs, forward schedules each one infinitely, and assigns a priority sequence.
2. Adjust Global Scheduling Order (optional) lets planners move jobs up or down the sequence.
3. Global Scheduling places every selected job on the schedule in that order, on the live schedule or a what-if schedule.

Run steps 1 and 3 together in a Process Set so the order is always recalculated first. Jobs that have been locked (for example on the Job Scheduling Board) keep their dates.

Pages in this section

• Implementing scheduling: the setup checklist, the nightly process set and the screens planners use
• Operation time: queue, setup, production and move
• Resources, resource groups and capabilities: how the engine chooses who does the work, and splitting setup from production
• Operation relationships and send-ahead
• Scheduling blocks and split operations

Related sections

• The job lifecycle: a job must be engineered before it can be scheduled

Sources

This page is written in our own words. It was informed by:

• EpiUsers: Job Entry rough cut scheduling