Glossary (EDGEBIC)

What Is an Operation State in a Production Gantt?

User Solutions TeamUser Solutions Team
|
6 min read

An operation state is a flag carried by one scheduled operation that records how it arrived at its current position, rendered on a scheduling Gantt as the color of that operation's bar. It is a provenance marker rather than a progress marker: it tells you whether the engine placed the operation, whether a planner dragged it, whether that drag has been saved, whether the operation ended up on a different machine than the engine chose, and whether an upstream change may have left it out of sequence. In EDGEBIC by User Solutions the full set of states is shown in a legend strip above the board, and each one can be recolored or switched off per site.

How it works

Every operation on the board starts in the default state, which simply means the scheduling engine placed it and nobody has touched it since. From there the state changes only in response to specific events, and the engine never invents one.

Editing moves it first. The moment a planner drags or resizes a bar, that operation flips to the override state. This is the single most informative state on the board because it is the only one that exists purely in memory: an overridden bar has not been written anywhere. Close the screen without saving and it snaps back. A board showing several overridden bars is a board with unsaved work on it.

Saving moves it again. When the pending set is written, each saved operation takes a state describing what the save actually recorded. An operation whose dates now differ from the engine's plan reads as applied. One that landed on a different work center reads as resource-replaced. One where only a start was recorded reads as start-applied, which is the in-progress case. One carrying a future planned pin rather than real actuals reads as planned-applied. And one whose recorded dates happen to match the engine's plan exactly reads as rescheduled, which is the common outcome immediately after a scheduling run.

One state is a warning rather than a record. When an earlier operation in a job is saved to a new position, the later operations in that same job are flagged for review. Nothing about them changed; the flag is saying that they may now be out of sequence relative to their new upstream end. It clears on the next reload after a re-plan.

Two treatments sit outside the state system entirely. A completed operation, or an operation belonging to a completed job, is painted with an opaque shade that covers whatever state color it had. The delivery tail after the last machining operation is drawn in its own muted treatment. Neither is a state, because neither describes how the operation was placed.

Because several of these conditions can be true at once, the states are resolved in a fixed priority order. Externally fed actuals outrank everything; a machine change outranks a date change; a future pin is evaluated ahead of the actual-based checks, since by definition a pin means no actuals exist yet.

A concrete example

A job runs saw, then mill, then paint. Monday morning the engine has planned all three and every bar is in the default state.

The mill was down Monday, so the planner drags the mill bar to start Tuesday at 08:00. That bar immediately turns to the override state and the pending count on the toolbar goes to one. Nothing has been written; a colleague looking at the same board on another machine still sees Monday.

The planner presses save. Three things now happen at once. The mill bar becomes applied, because it carries recorded dates that differ from the engine's plan. The paint bar, which is still sitting where the engine put it on Tuesday afternoon, is flagged for review, because the mill now ends later than paint was planned to start. The saw bar does not change at all, because nothing upstream of it moved.

The planner presses re-schedule. The engine re-plans only this job's remaining steps: paint moves to Wednesday morning and its review flag clears on reload. The mill's recorded start survives untouched, because recorded work is never re-placed.

Read as colors, that whole sequence is legible at a glance from across the room, which is the point. Read as a table of dates, it takes three columns and a comparison.

How EDGEBIC uses it

The state system is what makes the board's override-and-warn behavior safe to use. EDGEBIC will let a planner drag an operation almost anywhere and will warn rather than block, on the principle that the planner knows something the data does not. The states are the accounting that keeps that permissive behavior honest: an unsaved change is visibly unsaved, a machine substitution is visibly a substitution, and a knock-on effect on a downstream operation is visibly flagged rather than silently absorbed. How the colors map to the states on your own board, and how to change them, is covered in Gantt colors and labels in EDGEBIC.

Two states are worth learning by name because they carry the most operational weight. The override state marks the staged set that has not been written yet, described in how to batch your Gantt edits before saving in EDGEBIC. The review flag is the ripple indicator, described in how drag drop shows the downstream ripple, and the standard response to it is a targeted re-plan rather than a manual tidy-up.

Which state a save produces depends on what the save was told to record, and that is a separate setting rather than a property of the drag. See what is a drag write mode in a production gantt for the three things a saved drag can mean, and what is a planned start pin in scheduling for the state that expresses future intent without pretending work has happened.

The takeaway

An operation state answers how did this operation get here, and nothing else. Keeping that separate from how far along is this work removes most of the confusion a colored board creates, because the two questions genuinely have different answers and a single bar carries both. The practical habit is to scan for the override state before leaving the screen, since that is the only state whose information exists nowhere but the display in front of you. For the wider vocabulary see the manufacturing glossary, and for the product itself see EDGEBIC.

An operation state is a flag carried by a single scheduled operation that records how it arrived at its current position, and which is rendered as the color of that operation's bar on the Gantt. It is provenance rather than progress: it answers the question of who or what last moved this operation, not how far along the work is. That is why one bar can read as engine-placed while the bar beside it reads as manually saved, even though both are planned for the same afternoon and neither has started.

No, and confusing the two is the most common misreading of a scheduling board. Job status describes progress and lateness for a whole job. An operation state describes the editing history of one operation on one work center: whether the engine placed it, whether a planner dragged it, whether the change has been saved, whether it landed on a different machine than the engine chose, and whether an upstream save may have left it out of sequence. A completed job and a manually overridden operation are answers to different questions.

You can change the colors and you can switch individual states off, but not what each state signifies, because the meaning is what the scheduler records rather than a label you assign. The practical consequence is that you should always read a board against its own legend rather than against a screenshot from another site, since two plants running the same software can have completely different palettes for the same set of states.

Read the legend strip on the board first, because the color to state mapping is site-specific and the answer is usually right there. If the state turns out to be the manual-override one, someone dragged the bar and has not saved yet, and the change exists only on that person's screen. If it is the replaced-resource state, the operation was saved onto a different machine than the engine picked. For anything that needs a name and a timestamp rather than a color, open the audit trail for that job, which lists every move with its old and new dates.

Almost certainly not. A run that ends with a large block of operations sharing one state is normal, because a reschedule places many operations the same way at the same moment. The state to look at afterward is the review flag on operations whose upstream neighbor moved: those clear once the engine has re-planned around the change. If a review flag is still showing after a run, it usually means that operation was not in the run's scope, which is worth a second look rather than an alarm.

Expert Q&A: Deep Dive

Q: One of my bars is a color I do not recognize and nobody remembers setting it. How do I find out what happened?

A: Read the legend strip on the board first, because the color to state mapping is site-specific and the answer is usually right there. If the state turns out to be the manual-override one, someone dragged the bar and has not saved yet, and the change exists only on that person's screen. If it is the replaced-resource state, the operation was saved onto a different machine than the engine picked. For anything that needs a name and a timestamp rather than a color, open the audit trail for that job, which lists every move with its old and new dates.

Q: After a reschedule most of my bars turned the same color. Did something go wrong?

A: Almost certainly not. A run that ends with a large block of operations sharing one state is normal, because a reschedule places many operations the same way at the same moment. The state to look at afterward is the review flag on operations whose upstream neighbor moved: those clear once the engine has re-planned around the change. If a review flag is still showing after a run, it usually means that operation was not in the run's scope, which is worth a second look rather than an alarm.

Frequently Asked Questions

Ready to Transform Your Production Scheduling?

User Solutions has been helping manufacturers optimize their production schedules for over 35 years. One-time license, 5-day implementation.

User Solutions Team

User Solutions Team

Manufacturing Software Experts

User Solutions has been developing production planning and scheduling software for manufacturers since 1991. Our team combines 35+ years of manufacturing software expertise with deep industry knowledge to help factories optimize their operations.

Let's Solve Your Challenges Together