Gantt chart vs Kanban board for high-mix, low-volume production scheduling

We’re running a high-mix, low-volume operation with GPSF 2021.3 and debating whether to continue with Gantt chart scheduling or switch to the Kanban board interface for our shop floor. Currently using the standard Gantt view with about 40-60 active work orders daily across 12 production cells.

Our operators find the Gantt chart overwhelming - too much information and they struggle to identify which job to start next when multiple orders are scheduled for the same cell. We’re considering the Kanban board approach where operators just pull the next card from their queue.

Has anyone made this transition in a similar environment? I’m particularly interested in how the scheduling method impacts operator usability and overall throughput. Do you lose visibility into schedule conflicts and resource constraints with Kanban, or does the simpler interface actually improve shop floor efficiency?

That’s a valid concern. We do have some material dependencies, though our ERP system handles most of that upstream. The bigger question is whether we can configure the Kanban board to show material status or other constraints before operators pull a job. Anyone know if GPSF supports conditional card availability in the Kanban view?

We made this exact switch last year for a client with similar volumes. Kanban worked much better for operator usability - they stopped overthinking the schedule and just focused on pulling the next job. Throughput actually increased by about 12% because operators spent less time navigating the Gantt interface and more time on value-added work.

I’d also consider a hybrid approach. Use Gantt for scheduling and planning (planners and supervisors need that visibility), but present a simplified Kanban view to operators on the shop floor terminals. GPSF supports dual interfaces - your planning team works in Gantt mode while operators see only their prioritized queue in Kanban mode. Best of both worlds and you don’t lose the scheduling intelligence.