产能规划与运营瓶颈审查
结合需求波动、排队和停机损耗评估产能,输出运营瓶颈、情景测试及改进验证计划。
完整提示词
I want a deep, evidence-first and operations-oriented analysis that distinguishes symptoms from real constraints and local optimization from improvement of the total system. Main objective: > Determine whether planned capacity can absorb realistic demand, variability, maintenance, quality loss and growth without excessive queues, service degradation or economically wasteful overbuild. Do not do: - a generic lean checklist - automatic inventory or headcount reduction - treating high utilization as good without queue analysis - mixing local productivity with end-to-end throughput - cost cutting without service and resilience consequences - inventing missing operational data ## 1. CONTEXT DISCOVERY Establish: - operating model and customer promise - process boundaries and handoffs - throughput unit - demand profile, seasonality and variability - capacity by critical step - SLA/service level and quality standard - inventory, suppliers and lead times where present - critical technology and people dependencies - operating cost, margin and working capital - decision context and time horizon ## 2. EVIDENCE MODEL A - direct operational telemetry, ERP/WMS/MES/transaction log, timestamp, inventory count or production evidence B - complete process/data chain and multiple consistent sources C - strong derived analysis with transparent calculation D - credible inference requiring verification E - scenario or hardening recommendation Status: CONFIRMED / SUPPORTED / NOT VERIFIED / CONTESTED / NOT APPLICABLE / SCENARIO ## 3. METRIC AND FLOW DISCIPLINE For every key metric show: ```text Metric: Operational definition: Unit: Process boundary: Numerator: Denominator: Window: Source: Baseline: Variance: Target/constraint: Known exclusions: ``` Explicitly separate touch time, wait time, cycle time, lead time, throughput, capacity, utilization, yield, rework, service level and working-capital effect. ## 4. DEMAND FORECAST Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 5. PEAK DEMAND Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 6. ARRIVAL VARIABILITY Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 7. SERVICE VARIABILITY Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 8. EFFECTIVE CAPACITY Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 9. NAMEPLATE CAPACITY Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 10. UTILIZATION Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 11. QUEUEING EFFECT Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 12. MAINTENANCE Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 13. DOWNTIME Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 14. YIELD LOSS Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 15. SETUP TIME Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 16. SHIFT MODEL Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 17. LABOR CAPACITY Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 18. SKILLS CAPACITY Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 19. EQUIPMENT CAPACITY Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 20. SUPPLIER CAPACITY Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 21. CAPACITY BUFFER Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 22. EXPANSION LEAD TIME Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 23. CAPACITY ECONOMICS Analyze this area through the end-to-end flow, not an isolated local metric. For every material finding state the observed fact, process step, metric definition, baseline, capacity/variability context, evidence tier, customer effect, cash/margin effect, root cause, constraint interaction and what would falsify the finding. ## 24. CONSTRAINT AND QUEUE ANALYSIS For every potential bottleneck test whether queue grows before it, whether it operates near effective capacity, whether downstream starves, whether the constraint migrates by shift/day/SKU, and whether the issue is capacity, variability, setup, quality or scheduling. ## 25. FALSE-POSITIVE PROTECTION Do not report a problem merely because utilization is below 100%, safety stock exists, a supplier is single-source, a process has a manual step or capacity has buffer. A finding requires a real mechanism of harm and an end-to-end consequence. ## 26. SEVERITY P0 - near-term safety/regulatory shutdown, total delivery interruption or critical cash impact P1 - critical constraint that can stop core operations or break key customer commitments P2 - material throughput, quality, service, working-capital or margin issue P3 - limited efficiency or control gap P4 - hardening, monitoring or future optimization ## 27. FINDING FORMAT ```text ID: Severity: Status: Evidence tier: Process boundary: Observed condition: Metric: Baseline: Constraint: Trigger: Mechanism: Queue/flow effect: Quality effect: Customer effect: Cost/margin effect: Cash/working-capital effect: Dependencies: Root cause: Alternative explanation: Evidence: What would falsify this: Remediation: Verification: Owner: Decision trigger: ``` ## 28. REQUIRED MATRICES ### Flow Matrix | Step | Input | Capacity | Throughput | Queue/WIP | Yield | Lead time | Constraint | |---|---|---:|---:|---:|---:|---:|---| ### Dependency Matrix | Process/outcome | Dependency | Concentration | Failure signal | Recovery option | |---|---|---:|---|---| ### Economics Matrix | Driver | Volume | Unit cost | Variable/fixed | Cash timing | Margin impact | |---|---:|---:|---|---|---:| ### Control Matrix | Risk | Preventive control | Detective control | Owner | Evidence | Gap | |---|---|---|---|---|---| ## 29. FAILURE CHAINS Model at least 5 concrete chains: ```text condition ↓ trigger ↓ queue / capacity / quality / supply failure ↓ service or delivery failure ↓ cash / margin / customer consequence ``` Always test at least one combined scenario with two simultaneous shocks. ## 30. SCENARIO AND SENSITIVITY TESTS When relevant test: - demand +20% / +50% - lead time +25% / +50% - one critical supplier unavailable - yield or quality deterioration - 15% less effective capacity - slower replenishment and higher expedite cost - two incidents at once ## 31. ADVERSARIAL SECOND PASS - check whether local optimization worsened the system - look for hidden queues or off-system spreadsheets - inspect exception flow, not only the happy path - inspect peak periods, not only averages - test dependency failure and recovery time - check whether cost savings increase stockout, quality or continuity risk ## 32. FINAL QUALITY GATE Confirm process boundary, throughput unit, capacity, queue/WIP, variability, service level, quality, cost, cash, dependencies, false-positive protection, combined scenario, owner and verification step. ## 33. OUTPUT `CAPACITY_PLANNING_AUDIT.md` The report begins with Executive summary, Operating model, Top constraints, Customer/service risk, Cost/cash impact, Findings, Matrices, Failure chains, Improvement sequence and Verification plan. # FINAL RULE Do not optimize a local metric if it worsens end-to-end flow. The most important result is reliable customer outcome with sustainable throughput, quality, cost, cash and resilience.
分类与标签
商业营销、产能规划、运营瓶颈、需求波动、排队分析、供应链管理