Scheduling Boat Building Around Molds, Cure Time, and Launch Dates

A boat shop's capacity is not measured in labor hours, it is measured in molds and cure cycles, and a schedule that ignores either one will promise more hulls than the shed can physically turn. EDGEBIC by User Solutions counts each mold as a finite unit, carries lamination cure as real elapsed time, and schedules every hull backward from its launch date so the hull that cannot make spring is flagged in January.

Marine builders share a problem with almost nobody else. The constrained resource is a large hollow tool that is occupied for days, and the second constraint is time in which nothing happens at all.

Molds Are the Capacity

Three molds of a model, a four-day layup and cure cycle, and a quarter with 60 working days gives you roughly 45 mold cycles. That is the number, and it does not change because sales sold more boats.

Model each mold as a unit of a mold work center. If a mold accepts one layup at a time, flag the work center as one job per day so it commits to a single hull rather than being packed with hours. The scheduler will not place more concurrent layups than molds exist, and hulls that cannot claim a mold slot are moved and flagged. That constraint is the whole plan: everything downstream is timed off the day a hull comes out of the mold.

If some molds are interchangeable across models, put them into a work center group and bind the layup step to the group. The engine expands the group at schedule time and places the hull on the mold that frees up soonest.

Cure Is Time, Not Labor

Lamination cure, gelcoat cure, adhesive cure, and paint flash all consume calendar without consuming a person. A scheduler that only counts labor hours will place the demold step the same afternoon the layup ended.

Put the cure on the routing as queue time, a shift-aware buffer that sits after the operation completes and before the successor may start. A sixteen-hour cure lands as sixteen real hours, so demold happens the correct morning. The general shape is described in what is move time in a routing for related non-labor time, and the cure pattern in a neighboring sector in composites cure oven scheduling.

If the cure runs in a shared oven or heated shed with limited slots, model that as its own work center with real units rather than as buffer time, so two hulls cannot occupy one oven.

Backward From the Launch Date

Customers do not order a boat, they order a boat for a date. Sea trials, a boat show, a charter season, a delivery slot at a marina. Forward scheduling from now tells you when a hull could finish. Backward scheduling from the launch date tells you when layup has to start.

The engine places each hull's steps backward from its delivery date onto finite capacity. Hulls that fit are confirmed. Hulls that do not are flagged with the step that could not be placed, usually the mold. See forward versus backward scheduling for the trade-off and backward scheduling for make-to-order shops for the sold-to-order pattern.

Every Hull Is a Little Different

Even a production model carries options: a different helm package, an upgraded engine, custom joinery. Give the hull its own routing, or copy the base model routing and adjust the steps that genuinely change.

Once a hull is released to build, its routing is captured as a snapshot. An engineering change to the master model routing in March does not rewrite a hull that has been in the shed since January. That protection is described in how a frozen routing snapshot protects a running job, and the make-to-order context in engineer-to-order.

Specialized Crews Are a Constraint Too

Rigging, finish carpentry, and systems installation are done by small crews with real expertise, and a schedule that treats every bay as interchangeable will place work nobody available can do.

Model the specialization as a skill, assign it to the people who hold it, and require it on the routing step. The scheduler places that step only in shifts where a qualified person is rostered. You can pin a specific hull to a specific person when a customer asks for it, and that pin survives every later reschedule. When someone is out, remove them from the roster and reschedule: only unfinished work moves, and any hull that can no longer make its launch date is flagged.

A Worked Quarter

Eleven hulls of one model sold for spring, three molds, four-day mold cycle.

The mold is not the problem this quarter. The rigger is. That is exactly the kind of answer a whiteboard never produces, because the whiteboard tracks the resource everyone worries about rather than the one that binds.

Tracking Build Progress

Crews log start, stop, and progress at a shop floor station, so a hull that took two extra days in the finish bay pushes only the work that has not started. Completed steps are never moved by a reschedule, so the hull that launched last week keeps its record intact while the rest of the quarter reflows.

Heritage in Marine and Defense Programs

User Solutions has built finite capacity scheduling since 1991, more than 35 years, including work for the US Navy, GE, BAE Systems, and Cummins across 33 locations. The lineage behind EDGEBIC, including the RMDB heritage, coordinated the Nimitz refit at more than 26,000 tasks, a marine program where physical position, cure, and crew availability all constrained the same plan. A yard building eleven hulls a quarter is that discipline at a scale one planner can hold.

Where to Start

Model your molds as real units, put cure on the routing as queue time, add the two or three crew skills that genuinely gate work, and schedule this season's sold hulls backward from their delivery dates. The first flagged hull is the conversation you want to have in January.

For fundamentals, what is production scheduling is the plain-language start and make-to-order scheduling covers the sold-to-order pattern. The industry fit guide maps neighboring sectors and EDGEBIC is the product hub. Ready to find out how many hulls your molds actually support? Contact US for a demo.