How to Choose an Optimizer Goal in EDGEBIC

To choose an optimizer goal in EDGEBIC by User Solutions, open the Optimizer tab in the Schedule Jobs window and pick a goal from the goal picker. The four goals are On-Time First, Fastest Overall Finish, Least Setup Time, and Fewest Changes. Each is an ordered list of priorities the optimizer compares one tier at a time, so the goal you pick decides what the proposed plan is willing to trade. The optimizer shows a side-by-side comparison and persists nothing until you click Accept, which makes trying a goal a no-risk exercise.

This recipe covers one specific task. For what the goals mean in depth and how the tiers rank, read EDGEBIC optimizer goals and presets explained, and to run a full optimization end to end see how to run and read an optimization.

Before You Start

• A current schedule to optimize. The optimizer measures its proposal against the plan the greedy engine already produces.
• A clear priority for this run: protecting due dates, clearing the backlog soonest, cutting changeover, or minimizing disruption.
• For the Least Setup Time goal, a sequence-dependent setup matrix in place, since that goal only appears once changeover data exists.

Step by Step

1. Open the Schedule Jobs window and select the Optimizer tab.
2. In the goal picker, choose the goal that matches your priority:
   • On-Time First to protect due dates above all else (the default).
   • Fastest Overall Finish to clear the backlog soonest.
   • Least Setup Time to minimize total changeover (shown only when a setup matrix exists).
   • Fewest Changes to keep the plan close to what the floor already runs.
3. Choose a time budget (10, 30, or 60 seconds).
4. Click Run.
5. Read the verdict, the KPI delta grid, and the changes-only move list.
6. Click Accept to persist, or Discard to walk away.

What Changes When You Save

Running a goal changes nothing in the database. The optimizer builds a complete alternative plan, scores it against the greedy baseline under your chosen goal, and holds the proposal in memory until you decide. On-Time First ranks weighted tardiness first, then late job count, then stability, then finish time. Fastest Overall Finish leads with makespan. Least Setup Time leads with total setup. Fewest Changes leads with stability against the committed plan. Because comparison is tier by tier, a goal never lets a lower priority erase a higher one.

Only Accept writes a plan, and it uses the same persist path as a normal scheduling run, complete with an audit record of the goal, the KPIs, and the search settings.

How to Check It Worked

Read the KPI delta grid. It shows the baseline and the proposed value for every measure, so the goal you chose should show the biggest improvement on the priority it leads with. Choose Least Setup Time and total setup hours should fall; choose Fastest Overall Finish and makespan should shrink. The verdict badge tells you what kind of result you got: the multi-run engine reports the best of the schedules it tried, while the mathematical solver can report a proven optimality gap. If a goal shows no improvement, the move list is empty and the baseline stands.

Common Mistakes

• Fewest Changes on a first schedule. With no committed plan to stay close to, the stability priority is noise and EDGEBIC falls back to on-time tiers. Use Fewest Changes on a mid-week rerun, not a fresh plan.
• Looking for Least Setup Time without a setup matrix. The goal is hidden until changeover data exists. Build the setup matrix first, then the goal appears.
• Expecting a goal to override due dates silently. Every goal except On-Time First still keeps tardiness in its tier list, usually just below the lead priority, so due dates are never ignored outright. Read the tiers rather than assuming a single objective.
• Treating the badge language loosely. The multi-run engine finds the best of many complete schedules and is guaranteed never worse than the baseline; only the mathematical solver reports a proven gap. Match the claim to the engine you ran.

Where to Go Next

Choosing a goal is one half of using the optimizer well; the other is controlling what it may and may not move. See how to keep the optimizer from moving a job, and for the full workflow read the EDGEBIC optimizer guide. The full platform is at EDGEBIC, with more recipes in the EDGEBIC How-To Library.