AI Engineer
Summary
Internship-style role at an automotive components manufacturer improving visual quality inspection on the shop floor: analyzing defect and rejection data with quality tools (Pareto, 5 Why, Fishbone, PFMEA, Poka-Yoke) and building a camera-based machine-vision inspection proof of concept using Python/OpenCV and platforms like Cognex or Keyence.
Department: Quality / Manufacturing Engineering
Internship Objective : The internship will focus on improving the effectiveness of visual quality inspection in a manufacturing environment. Interns will work on identifying gaps in the existing inspection process and exploring technology-based solutions to reduce the risk of defective parts reaching subsequent processes or customers.
Project 1: Reduction of Defective Part Outflow from 100% Visual Inspection
Problem Statement
Despite 100% visual inspection, there is a possibility of defective parts passing through the inspection process. The project aims to study the existing inspection system, identify potential failure points, and develop practical measures to improve inspection effectiveness.
Key Responsibilities
The intern will:
- Study and map the existing 100% visual inspection process.
- Analyze historical defects, rejection data, and instances of defect outflow.
- Identify possible causes of inspection misses, including process, method, workplace, part presentation, and human-factor related issues.
- Conduct defect categorization and Pareto analysis to identify critical defect types.
- Perform root-cause analysis using tools such as 5 Why, Fishbone/Ishikawa analysis, and PFMEA.
- Evaluate the adequacy of current inspection standards, defect samples, lighting, inspection sequence, cycle time, and operator instructions.
- Identify opportunities for Poka-Yoke/error proofing and standardized inspection.
- Develop and test improvement actions.
- Establish measurable criteria to evaluate improvement in defect-detection effectiveness.
- Document findings, recommendations, and proposed sustainable controls.
Expected Deliverables
- Current-state process map
- Defect and outflow analysis
- Root-cause analysis
- Identification of critical inspection gaps
- Improvement recommendations and trials
- Standardized inspection methodology, where applicable
- Before-and-after effectiveness assessment
- Final project report and presentation
Project 2: Camera-Based Automated Visual Inspection System
Project Statement
Develop or evaluate a camera-based visual inspection solution capable of monitoring the inspection/operation sequence and supporting verification, correction, defect detection, and real-time notification.
Key Responsibilities
The intern will:
- Study the existing manufacturing and visual inspection sequence.
- Map the required operation and inspection sequence.
- Identify critical product characteristics and visually detectable defects.
- Define camera positioning, lighting, image-capture requirements, and inspection conditions.
- Explore suitable machine vision / computer vision approaches for defect detection.
- Develop a proof of concept for:
- Operation sequence mapping
- Sequence verification
- Detection of missed or incorrect operations
- Defect identification
- Operator notification/alert
- Evaluate image-processing approaches using tools such as OpenCV, Python, MATLAB, Cognex, Keyence, or equivalent vision platforms, depending on available equipment.
- Test the solution using acceptable and defective part samples.
- Measure the system for detection accuracy, false acceptance, false rejection, and cycle time.
- Recommend requirements for potential shop-floor implementation and scalability.
Expected Deliverables
- Existing and proposed inspection process flow
- Camera/lighting and system concept
- Proof of concept for camera-based inspection
- Defined defect-detection logic
- Operation-sequence verification logic
- Alert/notification concept
- Validation results including detection accuracy and false accept/reject cases
- Cost-benefit and implementation recommendations
- Final technical report and demonstration
Candidate Profile
Educational Background
Students pursuing B.E./B.Tech/M.Tech in:
- Mechanical Engineering
- Mechatronics
- Automation & Robotics
- Electronics / Electronics & Telecommunication
- Computer Engineering / Computer Science
- Artificial Intelligence / Machine Learning
- Industrial or Production Engineering
Preferred Skills
- Basic understanding of manufacturing and quality processes
- Interest in machine vision, automation, digital manufacturing, and Industry 4.0
- Analytical and problem-solving skills
- Basic knowledge of quality tools such as 5 Why, Fishbone, Pareto and FMEA
- For Project 2, exposure to Python/OpenCV, image processing, AI/ML, PLC/automation, Cognex or Keyence vision systems would be an advantage
- Ability to work on the shop floor, collect data, conduct trials, and document findings
- Good communication and presentation skills
What the Intern Will Gain
- Hands-on exposure to real manufacturing quality challenges
- Understanding of automotive/manufacturing visual inspection systems
- Practical application of quality and root-cause analysis tools
- Exposure to machine vision and automated inspection technologies
- Opportunity to develop and validate a solution in an actual production environment
- Experience working with Quality, Production, Engineering, and Automation teams