Scribd
Upload
5 views2 pages

Predictive Analysis For Manufacturing Operations

The document outlines an assignment to create a RESTful API for predictive analysis in manufacturing, focusing on predicting machine downtime or production defects. It includes details on dataset requirements, model selection, API endpoints for data upload, training, and prediction, as well as technical specifications and evaluation criteria. The final deliverables include a GitHub repository with the codebase, a sample dataset, and a README file for setup instructions, due within three days of receipt.

Uploaded by

202211053
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
5 views2 pages

Predictive Analysis For Manufacturing Operations

The document outlines an assignment to create a RESTful API for predictive analysis in manufacturing, focusing on predicting machine downtime or production defects. It includes details on dataset requirements, model selection, API endpoints for data upload, training, and prediction, as well as technical specifications and evaluation criteria. The final deliverables include a GitHub repository with the codebase, a sample dataset, and a README file for setup instructions, due within three days of receipt.

Uploaded by

202211053
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Predictive Analysis for Manufacturing Operations

Objective

To assess your ability to:

1.​ Build a simple predictive analysis model for manufacturing data.


2.​ Create RESTful API endpoints to upload data, train a model, and make predictions.

Scenario

You are tasked with creating a RESTful API to predict machine downtime or production
defects using a manufacturing dataset. The API should include endpoints for basic operations
and return predictions for decision-making.

Assignment Details

1.​ Dataset:​

○​ Use any small manufacturing-related dataset (e.g., Machine Downtime or


Product Defects). Publicly available options include Kaggle or UCI ML
Repository. If unavailable, generate synthetic data with key columns like
Machine_ID, Temperature, Run_Time, Downtime_Flag.
2.​ Model:​

○​ Use a simple supervised ML model (e.g., Logistic Regression, Decision Tree).


○​ Train the model on provided or uploaded data to predict machine downtime or
product defects.
3.​ Endpoints:​

○​ Upload Endpoint (POST /upload): Accept a CSV file containing manufacturing


data (e.g., Machine_ID, Temperature, Run_Time).
○​ Train Endpoint (POST /train): Train the model on the uploaded dataset and
return performance metrics (e.g., accuracy, F1-score).
○​ Predict Endpoint (POST /predict): Accept JSON input (e.g.,
{"Temperature": 80, "Run_Time": 120}) and return the prediction (e.g.,
Downtime: Yes/No).
4.​ Output Format:​

○​ Predictions: Return results in JSON format.


○​ Example Response: { "Downtime": "Yes", "Confidence": 0.85 }.
5.​ Technical Requirements:​

○​ Use Python with Flask or FastAPI for the API.


○​ Use scikit-learn for the model.
○​ Test endpoints locally using Postman or cURL.
6.​ Documentation:​

○​ Include a README with:


■​ Instructions to set up and run the API.
■​ Example API requests and responses.

Evaluation Criteria

1.​ Functionality: Are the endpoints working as expected?


2.​ Relevance: Does the model provide meaningful predictions for manufacturing use
cases?
3.​ Code Quality: Is the code clear and well-structured?
4.​ Delivery: Can the project be set up and tested in minimal time?

Expected Deliverables

1.​ A GitHub repository containing:


○​ Codebase.
○​ A sample dataset.
○​ README file with setup and usage instructions.
2.​ A working API that can be tested locally.

Deadline

Submit within 3 days from assignment receipt.

Good luck, and we look forward to seeing your implementation!

You might also like