Experiment: Performing ABC Analysis to Classify Inventory Items by Importance

🎯 Objective:

To classify inventory items into A, B, and C categories based on their annual consumption value, and analyze how focusing control on ‘A’ items can improve inventory efficiency.

---

🧩 Background Theory:

• ABC analysis is based on the Pareto principle (80/20 rule):

  • “A” items = ~20% of items contributing ~80% of total value

  • “B” items = ~30% of items contributing ~15% of total value

  • “C” items = ~50% of items contributing ~5% of total value

It helps inventory managers prioritize control, forecasting, and stock review.

---

🧮 Given Data (for Experiment):

Create a sample dataset of 10–15 inventory items with:

Item	Annual Demand (units)	Unit Cost (USD)
I1	1,000	10
I2	500	50
I3	300	100
I4	100	200
I5	1,200	5
I6	800	20
I7	150	80
I8	600	15
I9	100	300
I10	400	25

(You can randomize or expand this for larger experiments.)

---

🧭 Experimental Procedure:

1. Step 1 — Calculate Annual Consumption Value (ACV):
   
   

2. Step 2 — Rank Items by ACV:
   Sort the list in descending order (highest to lowest ACV).

3. Step 3 — Compute Cumulative % of Total Value and Items:

4. Step 4 — Classify Items:

   • Class A: Cumulative value up to ~80%

   • Class B: Next ~15%

   • Class C: Remaining ~5%

5. Step 5 — Plot ABC Curve (optional but powerful):
   Plot:

   • X-axis: Cumulative % of items

   • Y-axis: Cumulative % of value

   → The curve will show steep rise for A items and flat for C items.

6. Step 6 — Analysis & Decision:

   • How many items fall in A/B/C groups?

   • What % of value does each represent?

   • Suggest control policies:

     • A items: tight control, frequent review, accurate forecasting

     • B items: moderate control

     • C items: simple periodic review, bulk orders

---

📈 Expected Results:

• Roughly 2–3 items (20%) in Class A contribute ~80% of total value.

• 3–4 items (30%) in Class B contribute ~15%.

• Remaining items (50%) in Class C contribute ~5%.

---

🔍 Evaluation / Discussion:

• Did the results follow the 80/20 rule?

• What control strategies would reduce total inventory cost?

• How would misclassification affect stockouts or overstock?

---

💡 Extension Ideas (Optional):

• Repeat using real company data.

• Combine with XYZ analysis (demand variability).

• Simulate reordering frequency and service levels by class.

Python Code by Chat GPT

# 🧪 ABC Analysis Experiment for Inventory Management

import pandas as pd

import matplotlib.pyplot as plt

# --- Step 1: Create Sample Inventory Data ---

data = {

    "Item": ["I1", "I2", "I3", "I4", "I5", "I6", "I7", "I8", "I9", "I10"],

    "Annual_Demand": [1000, 500, 300, 100, 1200, 800, 150, 600, 100, 400],

    "Unit_Cost": [10, 50, 100, 200, 5, 20, 80, 15, 300, 25]

}

df = pd.DataFrame(data)

# --- Step 2: Compute Annual Consumption Value (ACV) ---

df["Annual_Value"] = df["Annual_Demand"] * df["Unit_Cost"]

# --- Step 3: Sort by Annual Value Descending ---

df = df.sort_values(by="Annual_Value", ascending=False).reset_index(drop=True)

# --- Step 4: Compute Cumulative Percentages ---

df["Cumulative_Value"] = df["Annual_Value"].cumsum()

df["Cumulative_Value_Percent"] = 100 * df["Cumulative_Value"] / df["Annual_Value"].sum()

df["Cumulative_Item_Percent"] = 100 * (df.index + 1) / len(df)

# --- Step 5: Classify A, B, C ---

def classify(row):

    if row["Cumulative_Value_Percent"] <= 80:

        return "A"

    elif row["Cumulative_Value_Percent"] <= 95:

        return "B"

    else:

        return "C"

df["Class"] = df.apply(classify, axis=1)

# --- Display the ABC Table ---

print("=== ABC Analysis Results ===")

print(df[["Item", "Annual_Demand", "Unit_Cost", "Annual_Value",

          "Cumulative_Value_Percent", "Class"]])

# --- Step 6: Plot the ABC Curve ---

plt.figure(figsize=(8,5))

plt.plot(df["Cumulative_Item_Percent"], df["Cumulative_Value_Percent"],

         marker='o', linestyle='-', color='blue')

# Add grid and labels

plt.title("ABC Analysis for Graphene Inventory")

plt.xlabel("Cumulative % of Items")

plt.ylabel("Cumulative % of Value")

plt.grid(True)

# Annotate A, B, C zones

plt.axhline(y=80, color='green', linestyle='--', label='A/B boundary (80%)')

plt.axhline(y=95, color='orange', linestyle='--', label='B/C boundary (95%)')

plt.legend()

plt.show()