diff --git a/blockchain_integration/pi_network/PiPulse/ai/models/preprocessing.py b/blockchain_integration/pi_network/PiPulse/ai/models/preprocessing.py
new file mode 100644
index 000000000..5765d002d
--- /dev/null
+++ b/blockchain_integration/pi_network/PiPulse/ai/models/preprocessing.py
@@ -0,0 +1,22 @@
+import pandas as pd
+import numpy as np
+from sklearn.preprocessing import MinMaxScaler
+
+def preprocess_data(data):
+    # Handle missing values
+    data.fillna(data.mean(), inplace=True)
+
+    # Scale data using Min-Max Scaler
+    scaler = MinMaxScaler()
+    data[['cpu_usage', 'memory_usage', 'disk_usage']] = scaler.fit_transform(data[['cpu_usage', 'memory_usage', 'disk_usage']])
+
+    # Extract features and labels
+    features = data.drop(['label'], axis=1)
+    labels = data['label']
+
+    return features, labels
+
+def split_data(features, labels, test_size=0.2):
+    from sklearn.model_selection import train_test_split
+    X_train, X_test, y_train, y_test = train_test_split(features, labels, test_size=test_size, random_state=42)
+    return X_train, X_test, y_train, y_test