Ta Workbook Filler

Logistics Simulation Workbook Filler

Use this skill for workbook-driven logistics simulation inputs: station availability (TA), plant/manual loss, station cycle time (CT), safety-zone aggregation, simulation input tables, and theoretical line capacity.

Use the spreadsheets skill and bundled workspace runtime for .xlsx/.xls/.csv work. Preserve source files unless the user explicitly asks to modify them; write derived workbooks, logs, and notes separately.

Scope

Treat body-in-white, welding, paint, final assembly, logistics, warehouse, AGV/AMR, conveyor, machining, packaging, and other production systems as variations of the same workbook problem. Do not hard-code automotive body-shop assumptions unless the provided sources or user instructions support them.

Source language may be Chinese, English, German, or another major engineering language. Preserve original station/equipment names where traceability matters, and add normalized aliases only for matching and calculation.

The skill may be only this SKILL.md. Scripts and references are optional helpers, not requirements. If later revisions add, remove, or contradict earlier guidance, apply the newest user intent while preserving generally useful rules.

Core Outputs

Typical outputs include:

• A filled TA calculation workbook, either using a provided template or a newly created structure from a TA library.
• A station-level input workbook for simulation software, with fields such as Name, object path, object type, CT/proctime by model variant, availability, MTTR, and optional PL rows.
• A safety-zone or lockout-zone summary with station-to-zone mapping, station TA, station PL, zone TA, and zone PL.
• A theoretical capacity workbook by line and model variant, including equipment-only, plant-loss, safety-zone, and mixed-model scenarios.
• A mapping/audit sheet listing missing stations, ambiguous mappings, low TA values, untrusted assumptions, and data gaps.

Source Priority

Use all available evidence, but keep traceability clear.

1. Direct station availability workbook: use as the station TA source, then verify against equipment/process evidence.
2. Station-level equipment list: best source for counting equipment quantities.
3. Process description PPT/PDF/Word: use to identify station scope, equipment actions, manual operations, joins, transfers, and side inputs.
4. Gantt-style actual cycle-time workbook: use for CT, station scope validation, and sometimes equipment/action evidence.
5. Safety-zone/lockout-zone documents or user screenshots: use for zone membership and zone-level availability aggregation.
6. CAD/layout drawings: only use if a capable tool/user-provided screenshot makes the relevant station/equipment information observable; otherwise mark as unavailable evidence.

If none of the required sources exist, do not fabricate TA or CT. Create a gap list and ask for the missing evidence or explicit assumptions.

Availability Calculation

There are two common workflows.

Direct Availability Exists

When a workbook already provides station availability:

1. Map station names to simulation objects.
2. Convert percentages consistently: 0.9825 in Excel may mean 98.25%; input tables may require 98.25.
3. Validate suspicious values against equipment/process evidence. Treat very low station TA as a review trigger, not automatically wrong.
4. Keep equipment TA and plant/manual loss separate unless the user explicitly requests a combined value.

Availability Must Be Computed

When no direct station TA exists:

1. Identify station boundaries and variants from equipment lists, process evidence, and Gantt sheets.
2. Count only equipment that can fail or directly affects production: robots, welders, grippers, guns, fixtures with active functions, sensors, conveyors, lifts, APC/AMR interfaces, joining systems, glue systems, and active controls.
3. Do not count passive racks, labels, guards, stands, storage-only structures, brackets, or fixtures without active failure behavior unless the source TA library treats them as failure items.
4. Match local equipment names to the supplied TA library semantically, not by exact string only.
5. Use QTY and utilisation according to the TA workbook or library rules. If utilisation is unknown, use 100% only as an explicit assumption.
6. If a TA library lacks manual/plant-loss items or non-equipment loss categories, do not invent those factors. Mark them as out of scope or ask for loss assumptions.

TA Workbook Construction

If a dedicated TA template exists:

• Preserve its formulas, workbook structure, station blocks, formatting, and summary logic.
• Fill only the intended input cells unless the user asks for structural changes.
• Add or remove station blocks only when station count requires it, and validate formulas after the change.

If no dedicated TA template exists:

• Build a clear workbook from the supplied generic/customer TA library.
• Prefer one sheet per station or one repeated station block per station, plus a summary sheet.
• Include columns for equipment category, matched TA-library item, evidence/source, QTY, utilisation, item availability, station TA contribution, and notes.
• Add a summary with Line, Station, ST_TA, optional ST_PL, optional Machine Section / Safety Zone, MS_TA, MS_PL, and audit status.

Plant/Manual Loss

Keep plant/manual loss separate from equipment TA:

• ST_TA: station equipment availability.
• ST_PL: station-level plant/manual/non-equipment loss, when supported by the library or provided source.
• MS_TA: safety-zone or machine-section availability from grouped station TA.
• MS_PL: safety-zone or machine-section plant/manual loss from grouped PL factors.

Treat blank or 0 PL entries as “no PL factor supplied” unless the workbook defines a different convention. Use factor 1.0 for “no PL loss” in product calculations.

For simulation inputs, a separate _PL dummy station can represent plant/manual loss if the model uses lockout-zone behavior. Avoid creating multiple independent PL dummies for the same personnel group or same zone unless the user confirms that those losses should stack independently.

Cycle Time From Gantt Sheets

For Gantt-style cycle-time workbooks:

1. Identify model variants and station/sheet aliases before extracting values.
2. Prefer Begin + Duration over cached End when formulas or converted files may contain stale values.
3. If Begin + Duration crosses the takt/cycle boundary, split the interval and wrap to the beginning of the cycle.
4. For one station/model, merge overlapping time intervals and count the union, not the sum of every duration.
5. Combine repeated sections for the same station before computing CT.
6. For sections spanning multiple stations, use row-level station references when present; otherwise mark for review.
7. Keep model variants separate. If a station is absent for one model variant, do not silently copy the other model unless the user confirms the route.

Always create a CT audit sheet with station, model, computed CT, source file/sheet, merged intervals, rows used, and review notes.

Safety Zones / Lockout Zones

When safety-zone documents, screenshots, or user-confirmed mappings are available:

1. Map each station and PL dummy to a zone.
2. If a zone contains only station objects, zone TA is usually the product of those station TA factors.
3. If a zone contains a PL dummy, zone PL is usually the product of the relevant PL factors.
4. If the simulation software stops all objects in a zone when any member fails, explain that zone grouping changes stop behavior and statistics even when the product factor is numerically similar.
5. If actual model LockoutZone membership differs from the TA Summary Machine Section, prefer the actual model membership for simulation-theory comparison and keep the Summary membership as a separate audit view.

Simulation Input Workbook

When creating or filling an input workbook:

• Use the current model object list as the authority for object names and object paths.
• Preserve object type conventions such as Station, AssemblyStation, Robot, Source, and Drain.
• Assign station CT and TA to the object that actually carries the equivalent process in the model. Do not double-count robot CT/TA if robot actions were already folded into a station.
• Keep model variants as separate CT columns when routes or actual CT differ.
• Set MTTR only when requested or when the workbook convention is clear. Common conventions include seconds, minutes, or Excel time fractions; infer before writing.
• Create mapping checks for source stations that do not exist in the input workbook and input objects that do not map to a source station.

Theoretical Capacity

For each line and model variant:

1. Compute ideal JPH from the bottleneck station CT: Ideal JPH = 3600 / max_station_CT_seconds.
2. For mixed-model calculations, use an explicit mix assumption. A practical default is station-level weighted CT, then take the line bottleneck: Mixed CT = max(J5_CT * J5_share + V23_CT * V23_share + ...).
3. Calculate scenario factors separately:
   • No safety zone / no PL: product of station equipment TA.
   • No safety zone / with PL: product of station equipment TA and all applicable station PL factors; this is an independent-loss comparison even if the actual model uses zones.
   • With safety zone / no PL: product of zone equipment TA factors.
   • With safety zone / with PL: product of zone equipment TA and zone PL factors.
4. Report JPH for every scenario: Scenario JPH = Ideal JPH * scenario_availability_factor.

State limitations clearly: buffers, starvation, blocking, setup loss, batch sequencing, failure mode (SimulationTime, OperatingTime, ProcessingTime), overlapping failures, and actual lockout-zone logic can make simulation results differ from first-order theoretical capacity.

Audit Rules

Every derived workbook should include:

• Assumptions: mix ratios, cycle boundary, failure mode interpretation, PL convention, and safety-zone source.
• Mapping_Check: unmapped stations, duplicate mappings, side/variant aliases, and source stations missing from the model.
• Low_TA_Check: stations below the user’s threshold or obviously lower than neighboring stations.
• CT_Audit: only when CT was extracted from Gantt sheets.
• Source_Trace: source file/sheet/page references for non-obvious mappings.

Do not hide uncertainty. Mark unresolved items as 待确认 or Review.

References and Scripts

Read references/equipment_matching_rules.md when equipment naming or TA-library matching is ambiguous.

The scripts in scripts/ are examples of repeatable workbook automation. Treat them as references, not universal tools: inspect and adapt them to the current workbook schema, sheet names, units, and station naming rules before running.