fix minor bug on calculation

src/calculation_time_constrains/service.py

@@ -17,31 +17,74 @@ from src.overhaul_scope.service import get as get_scope
 from .schema import CalculationResultsRead
 
 
-def get_overhaul_cost_by_time_chart(overhaul_cost: float, days: int, decay_base: float = 1.1) -> np.ndarray:
+# def get_overhaul_cost_by_time_chart(overhaul_cost: float, days: int,numEquipments:int ,decay_base: float = 1.1) -> np.ndarray:
+#     if overhaul_cost < 0:
+#         raise ValueError("Overhaul cost cannot be negative")
+#     if days <= 0:
+#         raise ValueError("Days must be positive")
+        
+#     exponents = np.arange(0, days)
+#     cost_per_equipment = overhaul_cost / numEquipments
+#     # Using a slower decay base to spread the budget depletion over more days
+#     results = cost_per_equipment / (decay_base ** exponents)
+#     results = np.where(np.isfinite(results), results, 0)
+#     return results
+
+
+def get_overhaul_cost_by_time_chart(overhaul_cost: float, days: int, numEquipments: int, decay_base: float = 1.1) -> np.ndarray:
     if overhaul_cost < 0:
         raise ValueError("Overhaul cost cannot be negative")
     if days <= 0:
         raise ValueError("Days must be positive")
-        
+    
     exponents = np.arange(0, days)
-    # Using a slower decay base to spread the budget depletion over more days
-    results = overhaul_cost / (decay_base ** exponents)
+    cost_per_equipment = overhaul_cost / numEquipments
+    
+    # Introduce randomness by multiplying with a random factor
+    random_factors = np.random.normal(1.0, 0.1, numEquipments)  # Mean 1.0, Std Dev 0.1
+    results = np.array([cost_per_equipment * factor / (decay_base ** exponents) for factor in random_factors])
+    
     results = np.where(np.isfinite(results), results, 0)
     return results
 
-def get_corrective_cost_time_chart(material_cost: float, service_cost: float, days: int) -> Tuple[np.ndarray, np.ndarray]:
+# def get_corrective_cost_time_chart(material_cost: float, service_cost: float, days: int) -> Tuple[np.ndarray, np.ndarray]:
+#     day_points = np.arange(0, days)
+
+#     # Parameters for failure rate
+#     base_rate = 0.2  # Base failure rate per day
+#     acceleration = 2.4  # How quickly failure rate increases
+#     grace_period = 170  # Days before failures start increasing significantly
+
+#     # Calculate daily failure rate using sigmoid function
+#     daily_failure_rate = base_rate / (1 + np.exp(-acceleration * (day_points - grace_period)/days))
+
+#     # Calculate cumulative failures
+#     failure_counts = np.cumsum(daily_failure_rate)
+
+#     # Calculate corrective costs based on cumulative failures and combined costs
+#     cost_per_failure = material_cost + service_cost
+#     corrective_costs = failure_counts * cost_per_failure
+
+#     return corrective_costs, daily_failure_rate
+
+def get_corrective_cost_time_chart(material_cost: float, service_cost: float, days: int, numEquipments: int) -> Tuple[np.ndarray, np.ndarray]:
     day_points = np.arange(0, days)
 
     # Parameters for failure rate
-    base_rate = 1.2  # Base failure rate per day
-    acceleration = 11.2  # How quickly failure rate increases
-    grace_period = 90  # Days before failures start increasing significantly
+    base_rate = 0.2  # Base failure rate per day
+    acceleration = 2.4  # How quickly failure rate increases
+    grace_period = 170  # Days before failures start increasing significantly
 
     # Calculate daily failure rate using sigmoid function
     daily_failure_rate = base_rate / (1 + np.exp(-acceleration * (day_points - grace_period)/days))
 
+    # Introduce randomness in the failure rate
+    random_noise = np.random.normal(0.0, 0.05, (numEquipments, days))  # Mean 0.0, Std Dev 0.05
+    daily_failure_rate = daily_failure_rate + random_noise
+    daily_failure_rate = np.clip(daily_failure_rate, 0, None)  # Ensure failure rate is non-negative
+
     # Calculate cumulative failures
-    failure_counts = np.cumsum(daily_failure_rate)
+    failure_counts = np.cumsum(daily_failure_rate, axis=1)
 
     # Calculate corrective costs based on cumulative failures and combined costs
     cost_per_failure = material_cost + service_cost
@@ -203,7 +246,8 @@ async def create_calculation_result_service(
 
     overhaul_cost_points = get_overhaul_cost_by_time_chart(
         calculation_data.parameter.overhaul_cost, 
-        days=days
+        days=days,
+        numEquipments=len(equipments)
     )