Supply Chain Optimization and Analysis

Project Overview This project focuses on analyzing and optimizing supply chain operations using a dataset that includes information about product types, sales, stock levels, and other relevant factors. The primary objectives are to:

Analyze Data: Understand the relationships between various supply chain factors and their impact on revenue. Build Predictive Models: Use machine learning techniques to predict revenue based on input features. Optimize Costs: Apply linear programming to minimize costs under given constraints. Visualize Results: Create an interactive dashboard to visualize revenue and other key metrics. Dataset The dataset used for this project is a supply chain dataset with the following columns:

Product type: Type of product (e.g., cosmetics, haircare, skincare) SKU: Stock Keeping Unit identifier Price: Price of the product Availability: Availability status Number of products sold: Total products sold Revenue generated: Revenue generated from sales Customer demographics: Demographics of customers Stock levels: Current stock levels Lead times: Time taken from order to delivery Order quantities: Number of products ordered Shipping times: Time taken for shipping Shipping carriers: Shipping carriers used Shipping costs: Costs associated with shipping Supplier name: Name of the supplier Location: Location of the supplier Lead time: Time required for production Production volumes: Volume of products produced Manufacturing lead time: Time taken for manufacturing Manufacturing costs: Costs of manufacturing Inspection results: Results of product inspections Defect rates: Rates of defects in products Transportation modes: Modes of transportation used Routes: Shipping routes taken Costs: Overall costs Methods Data Preprocessing Handling Missing Values: Filled missing values in numeric columns with the mean of each column. Encoding Categorical Variables: Used LabelEncoder to convert categorical columns into numeric values for analysis. Exploratory Data Analysis (EDA) Correlation Heatmap: Generated a heatmap to visualize the correlation between numeric features, helping identify significant relationships.

python Copy code

Compute correlation matrix

correlation = data[numeric_columns].corr()

Plot correlation heatmap

plt.figure(figsize=(14, 12)) sns.heatmap(correlation, annot=True, cmap='coolwarm') plt.title('Correlation Heatmap') plt.show() Machine Learning Model Feature and Target Definition: Defined features and target variable (Revenue generated).

Model Training: Used RandomForestRegressor to predict revenue.

Model Evaluation: Assessed model performance using Mean Squared Error (MSE).

python Copy code

Define features and target

features = data.drop(['Revenue generated'], axis=1) target = data['Revenue generated']

Split the data

X_train, X_test, y_train, y_test = train_test_split(features, target, test_size=0.2, random_state=42)

Train the model

model = RandomForestRegressor(n_estimators=100, random_state=42) model.fit(X_train, y_train)

Predictions

y_pred = model.predict(X_test)

Evaluate model performance

mse = mean_squared_error(y_test, y_pred) print(f'Mean Squared Error: {mse}') Cost Optimization Linear Programming: Applied linear programming to minimize costs under constraints.

python Copy code from scipy.optimize import linprog

Example cost coefficients and constraints

c = [data['Costs'].mean()] # Cost per unit A = [[1]] # Constraints matrix (example) b = [data['Stock levels'].mean()] # Constraint value (example)

Solve the optimization problem

result = linprog(c, A_ub=A, b_ub=b, method='simplex')

print("Optimal Order Quantity:", result.x) print("Minimum Cost:", result.fun) Visualization Interactive Dashboard: Built a Dash application to create an interactive dashboard that visualizes revenue by product type.

python Copy code import dash import dash_core_components as dcc import dash_html_components as html from dash.dependencies import Input, Output import plotly.graph_objs as go

Initialize the app

app = dash.Dash(name)

app.layout = html.Div([ html.H1("Supply Chain Optimization Dashboard"), dcc.Graph(id='revenue-graph'), dcc.Interval(id='interval-component', interval=1*1000, n_intervals=0) ])

@app.callback(Output('revenue-graph', 'figure'), [Input('interval-component', 'n_intervals')]) def update_graph(n): fig = go.Figure() fig.add_trace(go.Bar(x=data['Product type'], y=data['Revenue generated'], name='Revenue by Product Type')) fig.update_layout(title='Revenue by Product Type', xaxis_title='Product Type', yaxis_title='Revenue') return fig

if name == 'main': app.run_server(debug=True) Key Findings and Results Correlation Insights: The heatmap revealed strong correlations between certain features, such as Price and Revenue generated. Model Performance: The Random Forest model provided insights into factors affecting revenue, with an evaluated MSE of [9550764.101039082]. Cost Optimization: The linear programming solution provided an optimal order quantity of [0.] with a minimum cost of [0.0]. Dashboard Visualization: The interactive dashboard allows users to explore revenue trends by product type, offering actionable insights for decision-making. Conclusion This project demonstrates the use of data analysis, machine learning, and optimization techniques to enhance supply chain operations. The interactive dashboard adds a layer of usability, enabling stakeholders to visualize key metrics and make informed decisions.