	In Browser
	StumbleUpon
	del.icio.us
	Google
	Google Buzz
	reddit
	LinkedIn

	Facebook
	Twitter
	Linkedin
	E-Mail

Generative AI > Google Gemini API > Purchase Order Automation Agent

Purchase Order Automation Agent

Author: Venkata Sudhakar

ShopMax India's procurement team currently raises purchase orders manually when someone notices stock running low. This reactive approach leads to stockouts during peak demand periods and over-ordering when sales slow. An automated PO agent that monitors stock levels daily, compares them to reorder points, and generates structured purchase orders when action is needed removes human latency from the restocking cycle.

This tutorial builds a Purchase Order Automation Agent using ADK and Gemini. The agent scans the inventory for SKUs that have hit their reorder point, calculates the required order quantity based on lead time and safety stock, and drafts a formal purchase order for each vendor.

The below example shows automated PO generation for ShopMax India low-stock products.

from google.adk.agents import LlmAgent
from google.adk.tools import FunctionTool
from google.adk.runners import Runner
from google.adk.sessions import InMemorySessionService
import datetime

# --- Inventory and vendor config ---
INVENTORY = {
    "SKU-1001": {
        "name": "Samsung 55-inch 4K TV",
        "current_stock": 8,
        "reorder_point": 15,
        "reorder_qty": 50,
        "safety_stock": 10,
        "daily_sales_avg": 5,
        "vendor": "Samsung India Pvt Ltd",
        "unit_cost": 45000,
        "lead_time_days": 7,
    },
    "SKU-1002": {
        "name": "Sony WH-1000XM5 Headphones",
        "current_stock": 22,
        "reorder_point": 20,
        "reorder_qty": 30,
        "safety_stock": 8,
        "daily_sales_avg": 3,
        "vendor": "Sony India Ltd",
        "unit_cost": 20000,
        "lead_time_days": 5,
    },
    "SKU-1003": {
        "name": "Apple iPad Air 11-inch",
        "current_stock": 5,
        "reorder_point": 12,
        "reorder_qty": 25,
        "safety_stock": 6,
        "daily_sales_avg": 2,
        "vendor": "Apple India Distributor",
        "unit_cost": 58000,
        "lead_time_days": 10,
    },
}

# --- Tool: scan for reorder triggers ---
def scan_reorder_triggers() -> dict:
    """Find all SKUs that have hit their reorder point.
    Returns:
        List of SKUs requiring purchase orders
    """
    triggers = []
    for sku, data in INVENTORY.items():
        if data["current_stock"] <= data["reorder_point"]:
            days_cover = round(data["current_stock"] / data["daily_sales_avg"], 1) if data["daily_sales_avg"] > 0 else 999
            triggers.append({
                "sku": sku,
                "name": data["name"],
                "current_stock": data["current_stock"],
                "reorder_point": data["reorder_point"],
                "days_cover": days_cover,
                "urgency": "CRITICAL" if days_cover < data["lead_time_days"] else "NORMAL",
            })
    return {"trigger_count": len(triggers), "triggers": triggers}

# --- Tool: generate PO for a SKU ---
def generate_purchase_order(sku: str) -> dict:
    """Generate a formal purchase order for a SKU.
    Args:
        sku: Product SKU to reorder
    Returns:
        Complete purchase order with vendor, quantities, and total value
    """
    data = INVENTORY.get(sku)
    if not data:
        return {"error": f"SKU not found: {sku}"}
    order_qty = data["reorder_qty"]
    total_value = order_qty * data["unit_cost"]
    expected_delivery = datetime.date.today() + datetime.timedelta(days=data["lead_time_days"])
    return {
        "po_number": f"PO-AUTO-{sku[-4:]}",
        "sku": sku,
        "product_name": data["name"],
        "vendor": data["vendor"],
        "order_qty": order_qty,
        "unit_cost": data["unit_cost"],
        "total_value": total_value,
        "expected_delivery": str(expected_delivery),
        "current_stock": data["current_stock"],
    }

# --- Agent setup ---
scan_tool = FunctionTool(func=scan_reorder_triggers)
po_tool = FunctionTool(func=generate_purchase_order)

agent = LlmAgent(
    name="po_automation_agent",
    model="gemini-2.0-flash",
    instruction="""You are a purchase order automation agent for ShopMax India.
1. Call scan_reorder_triggers to find which SKUs need restocking.
2. For each triggered SKU, call generate_purchase_order.
3. Present CRITICAL urgency POs first (stock will run out before delivery).
4. For each PO, output a clean summary:
   PO Number | Vendor | Product | Qty | Unit Cost | Total Value | Expected Delivery
5. Flag CRITICAL orders with an alert message.
Format currency in Rs with commas.""",
    tools=[scan_tool, po_tool],
)

session_service = InMemorySessionService()
runner = Runner(agent=agent, app_name="po_agent", session_service=session_service)

import asyncio

async def run_po(query: str) -> str:
    session = await session_service.create_session(app_name="po_agent", user_id="procurement")
    from google.genai.types import Content, Part
    msg = Content(role="user", parts=[Part(text=query)])
    result = ""
    async for event in runner.run_async(user_id="procurement", session_id=session.id, new_message=msg):
        if event.is_final_response() and event.content:
            result = event.content.parts[0].text
    return result

response = asyncio.run(run_po("Scan inventory and generate purchase orders for any SKUs that need restocking."))
print(response)

It gives the following output,

Purchase Order Generation - ShopMax India
3 SKUs scanned  |  3 reorder triggers found

!! CRITICAL - STOCKOUT RISK !!
SKU-1003 | Apple iPad Air | Days cover: 2.5 days | Lead time: 10 days
Stock will run out before delivery. Expedite or source from alternate supplier.

PO-AUTO-1003 | Apple India Distributor
  Product : Apple iPad Air 11-inch
  Qty     : 25 units
  Unit    : Rs 58,000
  Total   : Rs 14,50,000
  Delivery: 2025-04-17 (10 days)

PO-AUTO-1001 | Samsung India Pvt Ltd
  Product : Samsung 55-inch 4K TV
  Qty     : 50 units
  Unit    : Rs 45,000
  Total   : Rs 22,50,000
  Delivery: 2025-04-14 (7 days)  [CRITICAL - 1.6 days cover]

PO-AUTO-1002 | Sony India Ltd
  Product : Sony WH-1000XM5 Headphones
  Qty     : 30 units
  Unit    : Rs 20,000
  Total   : Rs 6,00,000
  Delivery: 2025-04-12 (5 days)  [NORMAL - 7.3 days cover]

Total PO value this run: Rs 43,00,000

Deploy this agent as a daily Cloud Run job that reads live inventory from BigQuery and submits approved POs directly to vendor portals via API integration. Add a human-in-the-loop step for POs above Rs 10 lakhs using ADK callback patterns - the agent pauses and sends a Pub/Sub notification to the procurement manager for approval before the PO is submitted.

Send your comments, suggestions or queries regarding this site to [email protected].