The Unfinished WorldWilbert Limson

The Factory of Constraints · Co-founder · industrial decision support

Deltara — making factory plans feasible

A low-waste cutting plan is not automatically a schedule a factory can run.

Role
Co-founder of Deltara Owner-reported
Relationship
Co-founder
Where
Indonesia — flexible-film manufacturing

Slitting optimizer demonstrated · APS proposed · MRP future

The situation

In a flexible-film plant, giant jumbo rolls are slit into narrower rolls for customer orders. Sales wants to keep delivery promises. PPIC needs a schedule that can actually run. Production needs stable machine runs.

Their goals collide whenever urgent orders, changing film specifications and finite machine capacity meet. The design problem is to make those physical and organizational constraints visible, so people can approve a plan that works.

People in the picture

  • Sales teams keeping customer promises
  • PPIC planners building schedules
  • Production running the machines

What I did

I co-founded Deltara. Its work with a flexible-film manufacturer centres on this planning problem, in three layers at three different stages: a slitting optimizer that has been demonstrated, a proposed planning and scheduling system, and a possible later phase for materials.

Which parts I personally designed, led or built is being documented, and the client is not named until permission is confirmed.

  • Optimization
  • Production planning
  • Systems architecture
  • Data integration

Decisions that shaped it

  1. Treat cutting and scheduling as one problem

    A pattern that minimises trim can still break child-station limits, core and width rules, or the sequence downstream. Slitting and scheduling may need to feed back into each other rather than run one-way.

    Documented

  2. Model the constraints people actually live with

    Finite machine capacity, changeovers, jumbo-roll availability and cooling, aging and transfer times all shape what is feasible — so they are first-class inputs, not afterthoughts.

    Documented

  3. Keep a human approval step

    The proposed system presents schedule options and delivery risk. Planners approve or override, and only approved plans are exported toward ERP-compatible execution templates.

    Proposed

What was made

proposed planning data flowproposederp ordersmiddlewaresql databaseslittingoptimizerfinitescheduleriteratehuman approvalerp templatefor executionno live erp connection is implied · a scoping-stage architectureconceptual redraw · not client material
Planning data flowConceptual illustrationProposedERP-derived order data passes through middleware into slitting and finite scheduling, then to human approval, and back as an ERP-compatible plan. Proposed architecture.
one jumbo roll · two cutting patternsdemonstrateda · lowest trimexceeds child-station count · not runnable on this slitterb · feasiblemore trim · respects stations, core, width and weight limitsthen the schedule must also work downstreamillustrative widths · not client data
Slitting pattern boardConceptual illustrationDocumentedA jumbo roll's width divided into child rolls. The low-trim pattern on top breaks a station limit; the feasible one below wastes a little more.
finite scheduling across machinesproposedcooling · agingline 1line 2line 3changeoverdue dateaging windowconceptual · no schedule computed · not deployed software
Schedule loomConceptual illustrationProposedMachines as threads on a loom: orders, changeovers, cooling and aging windows, due dates. Proposed scheduling logic — not deployed.
the tool proposes · people decideproposedoption ahigher delivery riskoption blower delivery riskplannerapprovesor overrideserptemplatesalesppicproductioneveryone sees the same picture · conceptual
Approval and overrideConceptual illustrationProposedTwo feasible plans with their delivery risk. A planner approves one or overrides it; Sales and Production see the same picture.
materials & inventory — a future questionfuture phaseavailabilitysafety stockreplenishoutside the confirmed initial scope · deliberately left unbuilt
Materials shelf — a future questionConceptual illustrationProposedAvailability, safety stock and replenishment. A possible later phase, deliberately drawn unfinished.

Where it stands

  1. Slitting / trim optimization

    Demonstrated

    Jumbo-to-child roll pattern choices with machine-specific feasibility checks.

    Demonstrated in meeting materials. No plant-scale savings are claimed.

  2. Advanced planning & scheduling (APS)

    Proposed / scoped

    Finite scheduling across machines with changeovers, availability and approval.

    A proposed implementation — not deployed software.

  3. Materials & inventory planning (MRP)

    Future phase

    Availability, safety stock and replenishment questions.

    A possible later phase, outside the confirmed initial scope.

Results

Not claimed yet. Results appear here only once they are documented — until then, this page shows what was designed, built or proposed.

Limits and honest caveats

  • No go-live, throughput gain or trim saving is claimed.
  • Working assumptions discussed during scoping (such as a frozen pre-production window and typical lead times) are not validated policies.
  • An indicative project timeline from the scoping document is not a committed delivery date.

In the Living Mind

This world feeds these lessons: