Update app.py

app.py

@@ -1,35 +1,71 @@
 # app.py
-import pickle
-import numpy as np
 import gradio as gr
+import math
+import pandas as pd
 
-# 🔴 HARD-CODED API KEY (NOT SECURE)
-API_KEY = "AIzaSyCmwfnlC7nF1XzG2ck4RBwDiwIpJ0WVZVU"
+def production_planner(total_production, machines_available, daily_capacity_per_machine, target_days=None):
+    """
+    Smart Production Planner:
+    - total_production: total units to produce
+    - machines_available: number of machines
+    - daily_capacity_per_machine: units a machine produces per shift
+    - target_days: optional, desired completion time
+    """
+    if machines_available <= 0 or daily_capacity_per_machine <= 0 or total_production <= 0:
+        return "All input values must be greater than 0", "", "", "", pd.DataFrame()
+    
+    capacity_per_day_one_shift = machines_available * daily_capacity_per_machine
 
-if not API_KEY:
-    raise RuntimeError("API_KEY is missing")
+    if target_days:
+        shifts_per_day = math.ceil(total_production / (capacity_per_day_one_shift * target_days))
+        days_required = target_days
+    else:
+        days_required = math.ceil(total_production / capacity_per_day_one_shift)
+        shifts_per_day = 1
 
-# Load model
-with open("logistic_regression_model.pkl", "rb") as f:
-    model = pickle.load(f)
+    production_per_shift_per_machine = daily_capacity_per_machine
+    total_daily_capacity = capacity_per_day_one_shift * shifts_per_day
 
-with open("feature_columns.pkl", "rb") as f:
-    feature_columns = pickle.load(f)
+    # Generate shift distribution table
+    production_remaining = total_production
+    schedule = []
+    for day in range(1, days_required + 1):
+        for shift in range(1, shifts_per_day + 1):
+            shift_production = min(production_remaining, machines_available * production_per_shift_per_machine)
+            schedule.append({
+                "Day": day,
+                "Shift": shift,
+                "Production This Shift": shift_production,
+                "Production per Machine": math.ceil(shift_production / machines_available)
+            })
+            production_remaining -= shift_production
+            if production_remaining <= 0:
+                break
+        if production_remaining <= 0:
+            break
 
-def predict(daily_capacity, demand, material_available):
-    X = np.array([[daily_capacity, demand, material_available]])
-    risk = model.predict_proba(X)[0][1]
-    return round(float(risk), 3)
+    schedule_df = pd.DataFrame(schedule)
 
+    return days_required, shifts_per_day, production_per_shift_per_machine, total_daily_capacity, schedule_df
+
+# Gradio interface
 demo = gr.Interface(
-    fn=predict,
+    fn=production_planner,
     inputs=[
-        gr.Number(label="Daily Capacity"),
-        gr.Number(label="Demand"),
-        gr.Number(label="Material Available"),
+        gr.Number(label="Total Production Required"),
+        gr.Number(label="Number of Machines Available"),
+        gr.Number(label="Daily Capacity per Machine"),
+        gr.Number(label="Target Completion Days (optional)", value=None)
+    ],
+    outputs=[
+        gr.Number(label="Estimated Days Required"),
+        gr.Number(label="Shifts Per Day Required"),
+        gr.Number(label="Production per Shift per Machine"),
+        gr.Number(label="Total Daily Production Capacity"),
+        gr.Dataframe(label="Shift Distribution Schedule")
     ],
-    outputs=gr.Number(label="Capacity Risk Score"),
-    title="ML Capacity Planning"
+    title="Smart Production Planner",
+    description="Enter production requirements, machine info, and optional target days to calculate days, shifts, and a full production schedule."
 )
 
 demo.launch()