Augmented Reality for Quality Inspection
Our AI-enhanced visual inspection improves quality control and reduces the chances of errors
What is the role of inspection in manufacturing? Why is it important?
Inspection is a critical part in the quality control process in manufacturing.
Quality inspection ensures that product quality and work steps performed meet the defined standards. Raw materials and components are often checked prior to the assembly process; products can also be inspected during and post production. Frontline workers measure, examine, and test whether work steps were carried out correctly, and that products and their components were assembled or maintained according to specifications.
End-of-line inspection
AI-enhanced visual inspection detects false parts, and with AR, guides workers to improve quality control.
Step Check helps frontline workers to perform end-of-line inspection cycles faster, with higher accuracy and less risk of errors. Step Check uses AR and AI to identify parts, classify the object’s state, and automatically learn from pass/fail decisions.
AR helps solve key inspection challenges
Slow inspection cycles impact production outcome
High-production outcome requires fast inspection cycles. Checking parts and objects manually without the right tools during an entire shift is very demanding for frontline workers. A worker must manually assess if parts are assembled correctly, wires are fully connected, or welding points are at the right location. Comparing real objects with paper-based manuals or 2D images is time-consuming, leading to slow inspection cycles and delays for customers.
Manual quality checks have higher risk of human errors
Frontline workers must assess parts with the naked eye and ensure high accuracy. Identifying false parts with traditional documentation is a challenge. Workers get used to repetitive tasks, get tired, and make mistakes. Inspection results are incorrect, faulty products delivered, and rework is needed. This results in loss of revenue and low customer satisfaction.
Traditional quality assessment solutions lack automated protocoling
Manually documenting inspection results is a time-consuming extra step susceptible to human error or paperwork getting lost. What's more, quality processes often require image feedback that adds one more step to the worker’s task list.
Benefits of using AR for quality inspection
Step Check helps frontline workers to perform end-of-line inspection cycles faster, with higher accuracy and less risk of errors. Step Check uses AR and AI to identify parts, classify the object’s state, and automatically learn from pass/fail decisions.
How it works – Step Check
Step Check is PTC's visual inspection solution powered by artificial intelligence (AI) and our computer vision technology. Step Check makes quality inspections more efficient by improving accuracy and reducing errors.
Step 1: Locate
Wayfinding takes the technician directly to where each step needs to be performed. Once at the step location, instructions tell the technician what to check or inspect. AI automatically lets the technician know if the current state is correct. Now technicians can quickly find a part and trigger the next step while AI clearly visualizes pass/fail to avoid errors.
Step 2: Perform
Step-by-step guidance clearly shows the technician exactly what they need to check. Videos, images, and text from existing documentation can be used to assist. This allows technicians to quickly perform steps in the correct order and only move onto the next step when the step state is documented.
Step 3: Validate
AI learning improves each time the step is performed to improve the automatic step validations. Indicators clearly show technicians the current pass/fail state. This process is one of the fastest and most accurate ways to perform inspections.
Step 4: Document
Automatic documentation collects and stores all inspection data for easy access. The protocol includes all pass/fail steps with time stamps and image feedback for easy reporting. This allows for automated proof of quality and fewer reporting errors from manual paper-based entries between systems.