Wafer-Fault-Detection

Problem Statement: Wafer Fault Detection

Description:

• In electronics, a wafer (also called a slice or substrate) is a thin slice of semiconductor, such as a crystalline silicon (c-Si), used for the fabrication of integrated circuits and, in photovoltaics, to manufacture solar cells. The wafer serves as the substrate(serves as foundation for contruction of other components) for microelectronic devices built in and upon the wafer.

• It undergoes many microfabrication processes, such as doping, ion implantation, etching, thin-film deposition of various materials, and photolithographic patterning. Finally, the individual microcircuits are separated by wafer dicing and packaged as an integrated circuit.

• Wafers are predominantly used to manufacture solar cells and are located at remote locations in bulk and they themselves consist of few hundreds of sensors. Wafers are fundamental of photovoltaic power generation, and production thereof requires high technology. Photovoltaic power generation system converts sunlight energy directly to electrical energy.

• The motto behind figuring out the faulty wafers is to obliterate the need of having manual man-power doing the same. And make no mistake when we're saying this, even when they suspect a certain wafer to be faulty, they had to open the wafer from the scratch and deal with the issue, and by doing so all the wafers in the vicinity had to be stopped disrupting the whole process and stuff anf this is when that certain wafer was indeed faulty, however, when their suspicion came outta be false negative, then we can only imagine the waste of time, man-power and ofcourse, cost incurred.

Summary Of Datasets:

• The dataset contains information about 1000 wafers, each of which has been inspected for defects. The features of each wafer include its X and Y coordinates, the type of defect (crack, pit, or scratch), and the size of the defect. The labels indicate whether the wafer is faulty or not faulty.

• The dataset can be used to train a machine learning model to predict whether a wafer is faulty based on its features. This could be used to improve the yield of semiconductor manufacturing processes.

Here are some additional details about the dataset:

• The data was collected from a semiconductor manufacturing plant.
• The data was collected using a microscope.
• The data was labeled by a human expert.
• The data is clean and well-formatted.
• The data is representative of the real-world problem of wafer fault prediction.

To run this project

1. pip install -r requirements.txt

2. python app.py

Home Page

3. Got to http://localhost:5000

• For train the model go to below link>>>

4. Go To http://localhost:5000/train

• For prediction of data go to below link

5. Go To http://localhost:5000/predict
6. Upload CSV file
7. Output will save as csv file

THANK YOU!