Skip to content

charlieviettq/awesome-agent-skill / .claude/skills/biz-supply-chain

"biz-supply-chain"

Verified

"Analyze supply chain operations using the SCOR model across Plan, Source, Make, Deliver, and Return processes. Use this skill when the user needs to optimize supply chain efficiency, evaluate supplier performance, improve logistics, or design an end-to-end supply chain strategy — even if they say 'our deliveries are slow', 'supply chain costs are too high', or 'we keep running out of stock'.".

FileKindSize
SKILL.mdskill md5.8 KB
examples/sample_scenario.mdmarkdown5.4 KB
references/inventory-models.mdmarkdown9.7 KB
references/scor-metrics.mdmarkdown10.7 KB

references/scor-metrics.md

View on GitHub
# SCOR Model Metrics Reference

SCOR 將指標分為三層:**Level 1(策略)→ Level 2(診斷)→ Level 3(追蹤)**。本文件聚焦在 Level 1 與 Level 2,涵蓋完整公式、產業基準與計算範例。

---

## 指標架構:Reliability / Responsiveness / Agility / Cost / Assets

SCOR 的五大屬性橫跨所有流程:

| 屬性 | 衡量什麼 | 主要代表指標 |
|------|---------|------------|
| **Reliability**(可靠度) | 你說到做到嗎? | Perfect Order Rate |
| **Responsiveness**(回應速度) | 多快完成訂單? | Order-to-Delivery Cycle |
| **Agility**(靈活度) | 能應對需求波動嗎? | Upside Supply Chain Flexibility |
| **Cost**(成本) | 供應鏈運作成本多少? | Total Supply Chain Cost |
| **Assets**(資產效率) | 資產利用率如何? | Cash-to-Cash Cycle Time |

---

## Process 1:Plan

### 1.1 Forecast Accuracy(預測準確率)

```
Forecast Accuracy = 1 - (|Actual - Forecast| / Actual)
```

**計算範例:**

| SKU | 預測量 | 實際量 | |Actual - Forecast| | FA per SKU |
|-----|-------|-------|-------------------|-----------|
| A01 | 1,000 | 1,150 | 150 | 87.0% |
| A02 | 500 | 420 | 80 | 81.0% |
| A03 | 2,000 | 2,050 | 50 | 97.6% |

加權平均 FA(以銷售量加權):
```
Weighted FA = Σ(Actual_i × FA_i) / Σ(Actual_i)
            = (1150 × 87.0% + 420 × 81.0% + 2050 × 97.6%) / (1150 + 420 + 2050)
            = (1000.5 + 340.2 + 2000.8) / 3620
            = 3341.5 / 3620
            = 92.3%
```

**產業基準:**
- 消費品(穩定品項):85–92%
- 電子產品(季節性):75–85%
- 時尚服飾:60–75%(天花板,因需求本質難預測)
- < 70% 視為警戒,需啟動 S&OP 改善

**常見陷阱:** 別用「平均絕對誤差(MAE)」取代 FA 向高層報告——MAE 沒有方向性,無法區分「短缺」與「積壓」。

---

### 1.2 Inventory Days(庫存天數)

```
Inventory Days (DOI) = Inventory Value / (Annual COGS / 365)
```

或等效地:
```
DOI = Inventory Value × 365 / Annual COGS
```

**三種分類計算(更有診斷價值):**

| 庫存類型 | 公式 |
|---------|------|
| Raw Material DOI | Raw Material Inventory / (Annual Material Cost / 365) |
| WIP DOI | WIP Inventory / (Annual COGS / 365) |
| Finished Goods DOI | FG Inventory / (Annual COGS / 365) |

**產業基準(Finished Goods):**
- 電商(快時尚):15–25 天
- 消費電子:20–35 天
- 工業製造:30–60 天
- 汽車零件:45–90 天

**警示:** DOI 飆高通常是 Forecast Accuracy 低的滯後指標,先修 Plan 再看庫存改善。

---

### 1.3 Inventory Turnover(庫存周轉率)

```
Inventory Turnover = Annual COGS / Average Inventory
```
(DOI 的倒數乘以 365)

---

## Process 2:Source

### 2.1 Supplier On-Time Delivery Rate(供應商準時交貨率)

```
Supplier OTD = On-Time Deliveries / Total Deliveries × 100%
```

「準時」的定義必須事先與供應商白紙黑字約定:

| 定義版本 | 說明 |
|---------|------|
| **Strict** | 僅限合約日期當天 |
| **Window** | 合約日期 ±1 天 |
| **Early** | 提早交貨不罰,遲交才扣分 |

建議採用 **Strict** 版本做內部追蹤,Window 版本做供應商 SLA。

**供應商評分卡範例:**

| 評分項目 | 權重 | 計算方式 | 本季得分 |
|---------|------|---------|---------|
| OTD Rate | 30% | 實際 OTD % | 88% → 26.4 |
| Defect Rate | 25% | 100% - Defect% | 98.5% → 24.6 |
| Lead Time Consistency | 20% | 1 - StdDev(LT)/Mean(LT) | 85% → 17.0 |
| Invoice Accuracy | 15% | Correct invoices / Total | 96% → 14.4 |
| Responsiveness | 10% | 問題回應 <48h 比率 | 90% → 9.0 |
| **Total** | 100% | | **91.4 / 100** |

分級:≥90 A 級(策略供應商)、75–89 B 級(改善計畫)、<75 C 級(尋求替代)

---

### 2.2 Supplier Defect Rate(供應商不良率)

```
Supplier Defect Rate = Defective Units Received / Total Units Received × 100%
```

**產業基準:**
- 汽車 Tier-1:< 50 PPM(parts per million)
- 消費電子:< 0.5%(5,000 PPM)
- 一般製造:< 1–2%

---

### 2.3 Total Landed Cost(完整採購成本)

採購決策不能只看 Unit Price:

```
Total Landed Cost = Unit Price
                  + Freight Cost
                  + Import Duty & Taxes
                  + Customs Brokerage
                  + Warehousing-in-Transit
                  + Inspection Cost
                  + Payment Terms Cost (if not Net-0)
```

**Payment Terms Cost 計算:**
```
Payment Terms Cost = Invoice Amount × Annual Cost of Capital × (Payment Days / 365)
```

範例:供應商 A 報價 $100/unit(Net 60),供應商 B 報價 $102/unit(Net 0),資金成本 8%:
```
A: $100 + $100 × 8% × (60/365) = $100 + $1.32 = $101.32
B: $102 + $0 = $102.00
```
→ 看似貴的 B 其實 Landed Cost 更高,但差距很小;若 A 的 OTD 更差,可能反而更貴。

---

## Process 3:Make

### 3.1 OEE(Overall Equipment Effectiveness,設備綜合效率)

```
OEE = Availability × Performance × Quality
```

三個因子分解:

```
Availability  = (Planned Production Time - Downtime) / Planned Production Time
Performance   = (Ideal Cycle Time × Total Units Produced) / Run Time
Quality       = Good Units / Total Units Started
```

**實際計算範例:**

- 每班計劃工時:8 小時 = 480 分鐘
- 計劃停機(換線、休息):30 分鐘 → Planned Production Time = 450 分鐘
- 非計劃停機:45 分鐘 → Run Time = 405 分鐘
- Ideal Cycle Time:0.5 分鐘/件
- 實際生產:750 件
- 良品:720 件

```
Availability  = (450 - 45) / 450 = 405 / 450 = 90.0%
Performance   = (0.5 × 750) / 405 = 375 / 405 = 92.6%
Quality       = 720 / 750 = 96.0%

OEE = 90.0% × 92.6% × 96.0% = 80.0%
```

**產業基準:**
- World-class 製造:≥ 85%
- 一般製造平均:60–70%
- < 60%:嚴重問題,需立即啟動 TPM

**OEE 診斷方向:**

| 哪個因子低 | 根因方向 |
|-----------|---------|
| Availability 低 | 設備故障、維護不足、換線時間過長 |
| Performance 低 | 操作員技能、材料供給不穩、速度損失 |
| Quality 低 | 製程參數、材料品質、設計問題 |

---

### 3.2 Production Schedule Adherence(生產計劃達成率)

```
PSA = Actual Output / Planned Output × 100%
```

不要與 OEE 混淆:PSA 衡量「計劃執行率」,OEE 衡量「設備效率」。PSA 低可能是因為計劃本身不合理(需求預測差),而非生產能力問題。

---

## Process 4:Deliver

### 4.1 Perfect Order Rate(完美訂單率)——SCOR 最重要的 Level 1 指標

完美訂單 = 同時滿足以下四個條件的訂單:

```
Perfect Order Rate = (Orders On-Time) × (Orders In-Full) × (Orders Damage-Free) × (Orders with Correct Documentation) / Total Orders × 100%
```

**乘法結構(不是加法)** 是關鍵——任一環節失敗就不算完美訂單。

**計算範例:**

| 條件 | 達成率 |
|------|-------|
| On-Time | 96% |
| In-Full | 98% |
| Damage-Free | 99% |
| Correct Docs | 99.5% |
| **Perfect Order Rate** | **96% × 98% × 99% × 99.5% = 93.1%** |

即使每個環節都在 96–99.5% 之間,完美訂單率只有 93.1%。這就是為什麼末端客戶體驗往往比企業內部認知的差。

**產業基準:**
- 電商領先企業:≥ 95%
- 一般 B2C:88–93%
- B2B 製造業:90–95%

---

### 4.2 Order-to-Delivery Cycle Time(訂單交付週期)

```
O2D Cycle = Order Receipt → Order Processing → Warehouse Pick/Pack → Carrier Pickup → In-Transit → Last-Mile Delivery
```

分段追蹤,而非只看總時間:

| 階段 | 說明 | 典型時間(電商) |
|------|------|----------------|
| Order Processing | 訂單確認、付款核對、ERP 建單 | 0–4 小時 |
| Pick & Pack | 揀貨、包裝、標籤 | 2–24 小時 |
| Carrier Pickup | 等待取件 | 0–24 小時 |
| In-Transit | 幹線運輸 | 12–72 小時 |
| Last-Mile | 配送至消費者 | 4–24 小時 |

**分段追蹤才能定位瓶頸**:若 O2D 偏長,先看哪個階段的 P95 時間最大,而非整體平均。

---

### 4.3 Fill Rate vs. Order Fill Rate(填充率)

兩個常被混淆的指標:

```
Line Fill Rate  = Order Lines Shipped Complete / Total Order Lines
Order Fill Rate = Orders Shipped Complete / Total Orders
```

Order Fill Rate 更嚴格——一張訂單中任一 line item 缺貨就不算完整出貨。

---

## Process 5:Return

### 5.1 Return Rate(退貨率)

```
Return Rate = Units Returned / Units Shipped × 100%
```

退貨分類對診斷更有價值:

| 退貨原因 | 典型比率(電商) | 處置方向 |
|---------|---------------|---------|
| 商品描述不符 | 20–30% | 修改商品文案、圖片 |
| 尺寸/規格不符 | 25–35% | 尺碼建議工具、詳細規格 |
| 品質問題 | 10–20% | 追溯至 Make/Source |
| 物流損壞 | 5–15% | 包材升級、物流商 SLA |
| 買家改變心意 | 15–25% | 退貨政策設計 |

**產業基準:**
- 服飾電商:25–40%(天然高,不必恐慌)
- 消費電子:5–15%
- 工業品:< 3%

---

### 5.2 Reverse Logistics Cost Rate

```
RL Cost Rate = Total Reverse Logistics Cost / Total Forward Logistics Cost × 100%
```

業界估計:退貨處理成本約為正向出貨成本的 **1.5–2.5 倍**,因為需要增加揀選、檢驗、再分類、維修或銷毀等工序。

---

## Level 1 Dashboard 完整基準彙整

| Process | 指標 | 世界級基準 | 警戒線 |
|---------|------|-----------|-------|
| Plan | Forecast Accuracy | ≥ 85% | < 70% |
| Plan | Inventory DOI | 視品類 | 超過品類均值 50% |
| Source | Supplier OTD | ≥ 95% | < 85% |
| Source | Supplier Defect Rate | < 0.5% | > 2% |
| Make | OEE | ≥ 85% | < 65% |
| Make | PSA | ≥ 95% | < 85% |
| Deliver | Perfect Order Rate | ≥ 95% | < 88% |
| Deliver | O2D Cycle | 視通路 | 超過競品均值 |
| Return | Return Rate | 視品類 | 超過品類均值 20% |

---

## 優先序:當多個流程同時低於警戒線

遵循 TOC(Theory of Constraints)邏輯——找出制約點,不要同時改善所有流程:

1. **計算每個流程與基準的 Gap %**(Current vs. Benchmark)
2. **評估 Gap 對客戶體驗的影響**(Deliver 的 Gap 直接影響客戶,Plan 的 Gap 是根因)
3. **改善順序**:根因 Process 優先(通常是 Plan),即使症狀出現在 Deliver

**決策矩陣:**

| 症狀 | 表面流程 | 真正根因(優先改) |
|------|---------|----------------|
| 常常缺貨 | Deliver | Plan(FA 低)或 Source(OTD 低)|
| 客戶等太久 | Deliver | Make(OEE 低→產能不足)|
| 庫存堆積 | Plan | Plan(FA 低,預測偏高)|
| 退貨率高(品質原因) | Return | Make 或 Source |
| 總成本高 | Cost | 需跨流程分解,逐一排查 |
brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).Stars: 323kUpdated Claude CodeCodexVerifiedInstall
brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).Stars: 198kUpdated Claude CodeCodexVerifiedInstall