Scheduling Gear Manufacturing Across Hobbing, Heat Treat, and Grinding

A gear shop loses its throughput in two places: the changeovers on the hobbers and the week its parts spend at an outside heat treater. EDGEBIC by User Solutions models the hob changeover as a real sequence-dependent cost, carries the heat treat window as real elapsed time on the routing, and pools the grinders so finishing work flows to whichever machine is free.

The machining itself is rarely the problem. The problem is the order the machining happens in and the gap in the middle of the routing that nobody schedules.

Setups Are Not One Number

A hobber changeover depends entirely on what ran before. Two gears sharing a hob, arbor, and fixture may need fifteen minutes. A different module, a different bore, and a different cutter can cost three hours of teardown and setup. Averaging both into one setup time on the routing step is wrong in both directions: the easy transition is over-quoted and the hard one blows up the day.

Build a per-machine setup matrix. Group your gears into cutter and fixture families, then record the changeover hours from each family to each other family. Eight families gives you sixty-four cells rather than thousands of part-to-part entries, which is the difference between a maintainable table and a project nobody finishes. See how setup families reduce scheduling complexity and sequence-dependent setup times.

With real numbers in place, the scheduler adds the actual transition cost to the job it applies to, and the sequence optimizer can group compatible jobs on the constrained hobber. The multi-run layer is guaranteed never to return a schedule worse than the baseline it started from, so running it is free of downside: you compare total changeover hours and late-job count and accept only if it wins.

The Heat Treat Gap

Carburize, harden, temper, or nitride at an outside vendor is a week your parts are gone. Most schedules pretend that gap does not exist, so green machining looks like it feeds grinding directly, and the grinding department is idle Monday and drowning Thursday.

Put the outside process on the routing with its real transit and process window. A five-day carburize with a day each way of freight becomes seven days between green machining and hard finishing. Now green machining has a truck date attached, the grinding schedule is timed off the actual return, and the Thursday pileup becomes visible on Monday rather than at 6 AM Thursday. The related in-house pattern is covered in heat treating furnace scheduling.

If you use two vendors with different turnarounds, define them as alternates on the step with their own factors, and the engine picks whichever returns the parts earlier.

Pooling the Grinders

Gear grinding is expensive capacity and the routing usually names one machine because someone wrote it that way years ago. Every job then queues behind every other job on that grinder while an identical machine sits open.

Bind the grinding step to a work center group instead. At schedule time the engine expands the group into its members, compares each member's projected finish against live load, and places the job on the one that gets it done first. Members carry their own cycle time, so a newer machine that grinds a part in 0.8 of the standard is modeled honestly. Jobs already running stay on the machine running them, and only work that has not started re-shops the pool.

If one grinder is the only one that can do a particular module or diameter, leave that job bound to the machine and pool only what is genuinely interchangeable.

Inspection as a Real Step

Gear inspection on an analytical checker takes real time and there is usually one machine. Model it as a work center with its own capacity, not as an assumption. On a shop where every job routes through one checker before shipping, that machine is often the true constraint and nobody has ever measured it. Flag it as your bottleneck and the schedule anchors around it: how to flag and schedule around a bottleneck.

A Worked Job

A 200-piece helical gear order through a full routing.

The hob step landed on Hobber 2 because the prior job shared its cutter family, saving nearly three hours of changeover. Grinding landed on whichever grinder was free, not on the one written in the routing. The vendor window is on the calendar, so nobody is surprised when the parts are not back Monday.

Tracking What Really Happened

Operators log start, stop, and quantity at a shop floor station, so a hob run that took two extra hours pushes only the work that has not started. Completed operations are never moved by a reschedule, so the record of what ran stays intact through every replan. Over a few months the logged changeover times become the evidence your setup matrix needs to get more accurate, which is the loop that makes the sequencing gain compound.

Heritage in Precision Machining

User Solutions has built finite capacity scheduling since 1991, more than 35 years, for shops where machine time is the product: US Navy, GE, BAE Systems, and Cummins across 33 locations. The lineage behind EDGEBIC, including the RMDB heritage, drove GE Railcar on-time delivery from 30 percent to 90 percent in an environment of long routings and outside processes. A gear shop losing a week to a heat treater and a shift a day to changeovers is exactly that problem.

Where to Start

Pick your busiest hobber. Group the parts that run on it into four or five cutter families, fill in the changeover hours between them, put your real heat treat window on the routings that use it, and pool the grinders. Then reschedule and compare total changeover hours before and after.

For fundamentals, CNC machine scheduling covers the machining basics and job shop setup time reduction covers the economics. The industry fit guide maps neighboring sectors and EDGEBIC is the product hub. Ready to see how many changeover hours your due-date sort is spending? Contact US for a demo.