Update app.py

app.py

@@ -1,145 +1,173 @@
 import gradio as gr
 import pandas as pd
-import numpy as np
-
-PRIMARY_COLOR = "#0f2c59"
-
-# ==============================
-# WAREHOUSE MODEL PARAMETERS
-# ==============================
-NUM_AISLES = 30
-NUM_RACKS = 20
-
-FAST_ZONE = range(1, 6)     # aisles 1–5
-MID_ZONE = range(6, 16)     # aisles 6–15
-SLOW_ZONE = range(16, 31)   # aisles 16–30
-
-
-# ==============================
-# SLOT RECOMMENDATION ENGINE
-# ==============================
-def recommend_slot(velocity):
-    """
-    Assign aisle zone based on SKU velocity.
-    """
-    velocity = str(velocity).lower()
-
-    if velocity in ["fast", "high", "f"]:
-        aisle = np.random.choice(list(FAST_ZONE))
-        reason = "Fast-moving SKU — Assigned close to dispatch to reduce pick time."
-
-    elif velocity in ["medium", "mid", "m"]:
-        aisle = np.random.choice(list(MID_ZONE))
-        reason = "Medium-moving SKU — Assigned to middle zone to balance travel distance."
-
-    else:
-        aisle = np.random.choice(list(SLOW_ZONE))
-        reason = "Slow-moving SKU — Assigned to back aisles to avoid congestion."
-
-    rack = np.random.randint(1, NUM_RACKS + 1)
-
-    return aisle, rack, reason
-
-
-# ==============================
-# MAIN OPTIMIZATION FUNCTION
-# ==============================
-def optimize_slotting(df):
-    if df is None or len(df) == 0:
-        return None, "⚠️ Please upload SKU data."
-
-    results = []
-
-    for _, row in df.iterrows():
-        sku = row["SKU"]
-        velocity = row["Velocity"]
-        frequency = row.get("Frequency", "")
-
-        aisle, rack, reason = recommend_slot(velocity)
-
-        results.append({
-            "SKU": sku,
-            "Velocity": velocity,
-            "Frequency": frequency,
-            "Suggested Aisle": aisle,
-            "Suggested Rack": rack,
-            "Reason": reason
-        })
-
-    output_df = pd.DataFrame(results)
-    return output_df, "✅ Optimization Completed"
-
-
-# ==============================
-# GRADIO UI
-# ==============================
-def build_interface():
-
+import matplotlib.pyplot as plt
+import io
+import base64
+
+# ---------------------------------------------------
+# Helper: Generate Heatmap Image (Aisle × Rack)
+# ---------------------------------------------------
+def generate_slotting_heatmap(df):
+    plt.figure(figsize=(8, 6))
+    plt.scatter(df["Suggested Aisle"], df["Suggested Rack"],
+                c='orange', s=120, edgecolor='black')
+
+    plt.title("SKU Slotting Heatmap (Aisle × Rack)")
+    plt.xlabel("Aisle Number")
+    plt.ylabel("Rack Number")
+    plt.grid(True, alpha=0.3)
+
+    # Add SKU labels near points
+    for i, sku in enumerate(df["SKU"]):
+        plt.text(df["Suggested Aisle"][i] + 0.3,
+                 df["Suggested Rack"][i] + 0.3,
+                 sku, fontsize=9)
+
+    buf = io.BytesIO()
+    plt.savefig(buf, format="png", bbox_inches='tight')
+    plt.close()
+    buf.seek(0)
+
+    encoded = base64.b64encode(buf.read()).decode()
+    return "data:image/png;base64," + encoded
+
+
+# ---------------------------------------------------
+# Helper: Generate Before vs After Distance Chart
+# ---------------------------------------------------
+def generate_distance_chart(before_dist, after_dist):
+    plt.figure(figsize=(7, 5))
+
+    plt.bar(["Before", "After"], [before_dist, after_dist],
+            color=["red", "green"])
+
+    plt.title("Walking Distance Comparison")
+    plt.ylabel("Distance (meters)")
+
+    buf = io.BytesIO()
+    plt.savefig(buf, format="png", bbox_inches='tight')
+    plt.close()
+    buf.seek(0)
+
+    encoded = base64.b64encode(buf.read()).decode()
+    return "data:image/png;base64," + encoded
+
+
+# ---------------------------------------------------
+# Core Slotting Engine
+# ---------------------------------------------------
+def optimize_slotting(sku_df):
+    if sku_df is None or len(sku_df) == 0:
+        return None, "No data provided.", None, None
+
+    df = sku_df.copy()
+
+    # Mapping velocities to warehouse zone logic
+    velocity_zone_map = {
+        "Fast": (3, 14),      # near dispatch
+        "Medium": (6, 20),    # middle aisles
+        "Slow": (20, 6)       # far aisles
+    }
+
+    suggested_aisles = []
+    suggested_racks = []
+    reasons = []
+
+    for i in range(len(df)):
+        velocity = df.loc[i, "Velocity"]
+        freq = df.loc[i, "Frequency"]
+        sku = df.loc[i, "SKU"]
+
+        if velocity not in velocity_zone_map:
+            aisle, rack = (10, 10)  # fallback
+            reason = "Default placement due to unknown velocity."
+        else:
+            aisle, rack = velocity_zone_map[velocity]
+            reason = f"{velocity}-moving SKU — placed strategically to reduce picker travel."
+
+        suggested_aisles.append(aisle)
+        suggested_racks.append(rack)
+        reasons.append(reason)
+
+    df["Suggested Aisle"] = suggested_aisles
+    df["Suggested Rack"] = suggested_racks
+    df["Reason"] = reasons
+
+    # Business insight summary
+    insight = (
+        "��� **Business Insight Summary**\n"
+        "- Fast movers placed closest to dispatch to minimize walking distance.\n"
+        "- Medium movers placed in central aisles to balance traffic.\n"
+        "- Slow movers placed in deep aisles to avoid congestion.\n\n"
+        "**Operational Impact:**\n"
+        "✔ Walking distance reduction\n"
+        "✔ Faster order fulfillment\n"
+        "✔ Better warehouse throughput\n"
+        "✔ Lower aisle congestion\n"
+    )
+
+    # Generate visuals
+    heatmap_img = generate_slotting_heatmap(df)
+
+    before_distance = 300  # static demo number
+    after_distance = 180   # static demo number
+    distance_img = generate_distance_chart(before_distance, after_distance)
+
+    return df, insight, heatmap_img, distance_img
+
+
+# ---------------------------------------------------
+# UI Layout
+# ---------------------------------------------------
+def build_ui():
     with gr.Blocks() as demo:
-        gr.HTML("<style>.gr-button{background-color:#FF6A00 !important;color:white !important;}</style>")
-
         gr.Markdown(
-            "<h1 style='color:%s'>Procelevate Inventory Slotting Optimizer</h1>"
-            "<h3>AI-powered SKU placement engine to reduce picking time & congestion.</h3>"
-            % PRIMARY_COLOR
+            "<h1>Procelevate Inventory Slotting Optimizer</h1>"
+            "<p>AI-powered SKU placement engine to reduce picking time, congestion & improve throughput.</p>"
         )
 
-        gr.Markdown("### 📦 Upload SKU File or Enter Data")
+        gr.Markdown("📦 **Upload SKU File or Enter Data**")
 
-        template_df = pd.DataFrame({
-            "SKU": ["A123", "B555", "C888"],
-            "Velocity": ["Fast", "Medium", "Slow"],
-            "Frequency": [120, 60, 5]
-        })
-
-        sku_table = gr.Dataframe(
-            value=template_df,
+        # Input table
+        input_table = gr.Dataframe(
             headers=["SKU", "Velocity", "Frequency"],
-            row_count=5,
-            col_count=3,
-            wrap=True
-        )
-
-        optimize_btn = gr.Button("Optimize Slotting", size="lg")
-
-        result_table = gr.Dataframe(
-            headers=["SKU", "Velocity", "Frequency",
-                     "Suggested Aisle", "Suggested Rack", "Reason"],
-            row_count=5
+            datatype=["str", "str", "number"],
+            row_count=3,
+            value=[
+                ["A123", "Fast", 120],
+                ["B555", "Medium", 60],
+                ["C888", "Slow", 5]
+            ],
+            label="SKU Velocity Table"
         )
 
-        business_insight = gr.Markdown("")
+        submit_btn = gr.Button("Optimize Slotting", variant="primary")
 
-        def run(df):
-            result, status = optimize_slotting(df)
-            if result is None:
-                return None, status
+        # Tabs for results
+        with gr.Tabs():
+            with gr.Tab("Optimized Slotting"):
+                output_table = gr.Dataframe(
+                    headers=["SKU", "Velocity", "Frequency", "Suggested Aisle", "Suggested Rack", "Reason"],
+                    label="Optimized Slotting Table"
+                )
 
-            # Business insight summary
-            fast = len(result[result["Velocity"].str.lower() == "fast"])
-            medium = len(result[result["Velocity"].str.lower() == "medium"])
-            slow = len(result[result["Velocity"].str.lower() == "slow"])
+            with gr.Tab("Heatmap Visualization"):
+                heatmap_view = gr.Image(label="SKU Slotting Heatmap")
 
-            insight = f"""
-### 📊 Business Insight Summary
+            with gr.Tab("Walking Distance Savings"):
+                distance_view = gr.Image(label="Walking Distance Chart")
 
-- **Fast Movers:** {fast} SKUs moved closest to dispatch for quicker turnaround.  
-- **Medium Movers:** {medium} SKUs distributed in central aisles to balance traffic.  
-- **Slow Movers:** {slow} SKUs placed in far aisles to reduce congestion.  
+            with gr.Tab("Business Insights"):
+                insight_box = gr.Markdown()
 
-This improves:
-- 🚶 **Walking distance reduction**
-- 🚚 **Faster order fulfillment**
-- 🎯 **Better space utilization**
-- 🔄 **Lower aisle congestion**
-            """
-
-            return result, insight
-
-        optimize_btn.click(run, inputs=[sku_table], outputs=[result_table, business_insight])
+        submit_btn.click(
+            optimize_slotting,
+            inputs=[input_table],
+            outputs=[output_table, insight_box, heatmap_view, distance_view]
+        )
 
     return demo
 
 
-demo = build_interface()
+demo = build_ui()
 demo.launch()