Validation & Verification of IoT Applications for an Industrial Vision Sensor Technology Provider
About the customer
The customer is a leading provider of industrial automation solutions specializing in sensor technologies, machine vision systems, and factory automation.
Their products are used by OEMs, machine builders, robotics integrators, and manufacturing companies across a wide range of industries to automate quality inspection and defect detection.
The customer offered three different IoT cloud applications – Mobile, Web and Desktop apps to enhance the device management capabilities of the industrial vision sensor. These applications:
- Discovered and configured the industrial vision sensors.
- Reported results of the quality inspection.
Business challenge
Ensuring the apps delivered on seamless functionality & feature-rich user experience was a top priority.
Software defects, performance bottlenecks, or usability issues in the applications can interrupt inspection processes and potentially result in production downtime and poor ownership experience.
To avoid this, the customer required a structured validation process that could:
- Verify newly developed features updates & UI before release.
- Detect functional defects early in the development lifecycle.
- Validate application performance under production-like workloads.
- Improve the reliability and stability of milestone releases.
- Reduce repetitive manual validation effort through automation.
- Ensure consistent user experience across HMI and web applications.
Embitel’s solution
We supported the customer with comprehensive testing and validation services for the custom IoT Applications (Desktop, Mobile and Web) development for configuring industrial vision sensors.
Our validation team worked alongside the development teams to ensure every release met functional, performance, and quality expectations before deployment.
Requirement Analysis & Test Design
For every new feature, the customer provided user stories, functional requirements, and representative sample images through Redmine.
Embitel’s validation engineers:
- reviewed these requirements
- created detailed test cases
- validated the implemented functionality
and suggested usability improvements that could enhance the end-user experience.
End-to-End Validation Workflow
The team followed a standardized testing process for both HMI and Web applications consisting of:
- Sanity Testing
- Smoke Testing
- Password Protection & Security Validation
- Bug Fix Verification
- Software Task & Feature Testing
- System Testing
- Regression Testing
- Performance Testing
- Negative Testing
This consistent workflow ensured that unstable builds were identified early while validated builds progressed through subsequent testing stages.
Functional & Feature Validation
The validation team tested both newly introduced and continuously evolving features across multiple milestone releases. Brief examples of features tested include:
- Application Login Security
- Optical Character Recognition (OCR) using Machine Learning
- Sharing training model data across multiple locations
- Web layout and Code assist
Each milestone underwent feature verification before being approved for further testing.
Performance Validation
The team verified that machine vision applications continued to meet expected performance requirements under heavier workloads.
This was achieved by validating larger datasets and job configurations.
Performance bottlenecks were identified during development to prevent delays in production deployments.
Defect Identification & Quality Assurance
Testing was performed against every milestone build. Defects identified during validation were logged in Redmine, tracked through resolution, and revalidated after fixes were delivered by the development team.
The validation effort covered blocker issues, functional defects, performance issues, feature-related bugs, and password protection issues.
Test Automation
To improve validation efficiency, we implemented automated testing for repetitive validation activities.
Automation included:
- Python + Selenium for Web application testing.
- Python + Pywinauto for HMI application automation.
The team automated more than 100 web test cases and enabled parallel execution of routine smoke and sanity tests.
This reduced repetitive manual validation effort while allowing validation engineers to focus on higher-value testing activities.
Embitel’s impact
Through structured testing and validation, we helped improve the quality and stability of industrial machine vision application software before deployment.
Key outcomes of our validation & verification (V&V) efforts were:
- Achieved approximately 70% smoke test automation coverage across both Web and HMI.
- Time taken for manual testing was 2–3 hours. With automation, the execution took approximately 45 minutes or less.
- Automation of over 100 web test cases, reducing repetitive manual validation effort and freeing up time for feature testing and stability validation.
- Parallel execution of routine validation activities to improve testing efficiency.
Tools & technologies
- Tera Term
- Hercules
- PyCharm
- VS Code
- Wireshark
- Python
- Selenium
- Pywinauto
- pytest
- Redmine
