Skip to content
Skip to case study
Back to work
CASE STUDY / 02LinkedInX

Decision design · 2025

BizPsy Simulator

A strategy simulator that makes the link between consumer behavior, inventory and cash flow visible.

TL;DR

A simulator that models week-over-week demand elasticity and turns every decision into sales, remaining stock and revenue—with a financial reading written for non-analysts.

7 daysSimulated cycle per run
3 variablesPrice, elasticity and initial stock
0 jargonFinance — reading written for non-analysts
  • Weekly scenarios
  • Demand variables
  • Plain financial reading
PythonData modelsChart.js

01 / CONTEXT

The problem

Someone running a small business usually sets price, stock and promotions by intuition, without seeing clearly how those decisions chain into cash flow.

02 / DIRECTION

The solution

A simulator that models week-over-week demand elasticity and turns every decision into sales, remaining stock and revenue—with a financial reading written for non-analysts.

03 / EXECUTION

The process

01Architecture
02Prototype
03Iteration
04Delivery
  1. 01Research on simple price–demand elasticity models applicable to retail.
  2. 02Pure-Python prototype to validate the simulation engine before thinking about UI.
  3. 03Iterating the dashboard so the financial reading requires no finance vocabulary.
  4. 04Stress tests with extreme scenarios (highly elastic demand, critical stock) to verify stability.

04 / RESULT

Impact signals

7 daysSimulated cycle per run
3 variablesPrice, elasticity and initial stock
0 jargonFinance — reading written for non-analysts

05 / IMPLEMENTATION

A decision in code

simulator.py

import numpy as np

def simulate_weekly_cycle(demand_elasticity, base_price, stock):
    """Simulate user cashflow and inventory rates over a 7-day run."""
    print("[i] Loading demand variables for simulator...")
    days = 7
    daily_sales = []
    current_stock = stock
    
    # Calculate price reaction based on elasticity coefficient
    reaction = np.random.normal(1.0, 0.15, days)
    for day in range(days):
        demand = int(100 * (base_price ** -demand_elasticity) * reaction[day])
        sales = min(demand, current_stock)
        current_stock -= sales
        daily_sales.append(sales)
        
    revenue = sum(daily_sales) * base_price
    print(f"[BizPsy] Run done. Revenue: ${revenue:.2f} | Remaining stock: {current_stock}")
    return {"sales": daily_sales, "revenue": revenue, "final_stock": current_stock}

06 / LEARNING

Reflection

I learned that simplifying the math is easy; simplifying how you explain it without lying about the model's uncertainty is the hard part. I am still tuning that.