IC4004/31 Internship Assignment Written Report Sample 2026 | Singapore Polytechnic Singapore

1. Introduction

1.1. Name, History, Background of the company

ST Engineering was established in 1967 and has grown into one of Asia’s leading technologies, defence, and engineering groups. Headquartered in Singapore, the company operates across the aerospace, smart city, defence, and public security sectors, serving customers in more than 100 countries. With a strong focus on innovation and digital transformation, ST Engineering develops advanced solutions that enhance the safety, reliability, and efficiency of critical infrastructure.

1.2. Brief Details about the internship

This report covers my internship at ST Engineering from 16 March to 14 August 2026 in the Mobility, Rail & Road – Urban Solutions business. I was attached to the Rail Electronics team under the guidance of my supervisor, Chung Sudin, where I contributed to the Platform Screen Door Condition-Based Monitoring (PSD-CBM) system and other pro. My work involved software development, data analysis, system testing, dashboard development and technical documentation, providing valuable exposure to real-world railway engineering and intelligent transportation systems.

2. About the Company

2.1. Company Culture and Org Chart

2.1.1. Company Culture

Guided by its purpose of “Engineering a Better Future,” ST Engineering fosters a culture of innovation, integrity and collaboration. As one of Asia’s leading technology, defence and engineering groups, the company is committed to delivering smart and sustainable solutions that enhance safety, security and operational efficiency. Its core values emphasise customer focus, teamwork, excellence and continuous improvement, while encouraging employees to embrace innovation and lifelong learning. This collaborative and quality-driven culture enables employees to develop practical engineering solutions and contribute effectively to projects that support critical infrastructure and urban mobility systems.

2.1.2. Org Chart

ST Engineering is organised into several business areas, including Aerospace, Smart City, Defence & Public Security, and Commercial. My internship was under the Urban Solutions business, specifically within the Mobility, Rail & Road division. The detailed organizational chart is internal and confidential (Annex C).

2.1.3. Workflow

The workflow for process improvement is highly collaborative. Project requirements are first gathered from customers and system engineers before being translated into technical specifications. Software developers then implement new features or enhancements, followed by system integration and testing to verify functionality and reliability. Test results are documented, reviewed and refined through multiple iterations before deployment to the operational environment. Throughout the process, version control, technical documentation and collaboration among software engineers, test engineers and project managers ensure that project requirements are met while maintaining the reliability and safety of railway systems.

2.1.4. Interesting Features of the Company

Beyond its core operations, several features make ST Engineering a unique workplace:

• Global Engineering Group: Operating in over 100 countries, ST Engineering delivers engineering and technology solutions across multiple industries, providing employees with exposure to large-scale international projects.
• Culture of Innovation: The company encourages continuous improvement, innovation and the adoption of emerging technologies to address complex engineering challenges.
• Collaborative Workplace: Cross-functional teams comprising engineers, software developers and project managers work together to deliver integrated solutions for customers.
• Learning and Professional Growth: Employees are supported through structured training, mentorship and opportunities to develop both technical expertise and professional competencies.

2.2. My Work Experience (as an Intern)

2.2.1. The Intern’s Role and How it fits in the organization

At the Mobility, Rail & Road – Urban Solutions business, I was assigned to the Rail Electronics team under the supervision of Ms. Chung Sudin. The team is responsible for developing, testing and maintaining software solutions that support the monitoring, operation and maintenance of railway infrastructure.

My role was to provide direct support to the engineers on the team by contributing to software development, system testing, data analysis and technical documentation. I was involved in developing and enhancing the Platform Screen Door Condition-Based Monitoring (PSD-CBM) system, creating Grafana dashboards, generating test data, validating system functionality and preparing user documentation. By supporting these activities, I contributed to improving the reliability and efficiency of the system while gaining practical experience in real-world railway engineering projects.

2.2.2. The Project – CDIO Stages

During my internship, I was involved in a range of software development, system enhancement and automation projects within the Rail Electronics team. My responsibilities included developing internal tools, enhancing existing software modules, producing technical documentation and supporting system maintenance activities. These projects were carried out to improve engineering workflows, enhance system usability and increase operational efficiency.

This report focuses on four key contributions undertaken during my internship. These include the development of the CBM Test Data Generator and its accompanying User Manual, which were created to support the testing and validation of the Platform Screen Door Condition-Based Monitoring (PSD-CBM) system. The report also covers my enhancements to the NEA PDF Report Generator, which was developed for a separate project, as well as the development of a Database Backup Utility that enables users to export selected PostgreSQL and InfluxDB tables based on a specified date range. Each project will be discussed in the following sections, with the CBM Test Data Generator presented using the CDIO (Conceive, Design, Implement and Operate) framework.

A. CBM Test Data Generator (with User Manual)

Around the sixth week of my internship, I was introduced to the CBM Test Data Generator, a supporting tool for the Platform Screen Door Condition-Based Monitoring (PSD-CBM) system. The existing testing process required engineers to manually prepare and modify large amounts of test data, making system validation repetitive and time-consuming. Therefore, I was tasked with designing and developing a Test Data Generator that could automate the creation of test datasets while meeting the functional requirements of the engineering team. Throughout the development process, I worked closely with my supervisor, Ms Chung Sudin, who provided guidance, technical feedback and recommendations.

(C)  Conceive

1. Objective and Problem Definition: The primary objective was to develop a Test Data Generator to support the testing and validation of the Platform Screen Door Condition-Based Monitoring (PSD-CBM) system. By automating the generation of configurable test datasets, the application enabled engineers to perform functional testing more efficiently and consistently while significantly reducing the manual effort required for test preparation.

2.   Defining Functional Requirements:

a. Configurable Test Data Generation: Generate configurable and realistic datasets based on user-defined parameters to simulate various operating conditions and fault scenarios within the PSD-CBM system.

b. Database Compatibility: Produce test data compatible with PostgreSQL and InfluxDB, enabling direct use during system testing and validation.

c. User-Friendly Interface: Users needed to be able to review the generated records and download them as CSV

3. Technology Selection:

I selected Python because it provided suitable libraries for spreadsheet processing, data manipulation, simulation and file generation.

I used Streamlit to develop the interface. Streamlit allowed me to create an interactive, browser-based application using Python. It provided components such as date selectors, checkboxes, file uploaders, progress bars, editable tables and download buttons.

4. Defining Core Modules: To meet the project objectives, the application was designed with several key modules:

a. Parameter Configuration: Allow users to specify stations, assets, date ranges and testing parameters required for different validation scenarios.

b. Data Generation Engine: Generate realistic sensor data, alarm records and database entries based on predefined engineering rules and reference data.

c. Database Export: Produce datasets in formats compatible with PostgreSQL and InfluxDB for direct use during system testing.

(D) Design

With the concept and platform finalized, the design phase focused on the system’s architecture.

• System Architecture: The Test Data Generator was designed as a modular application instead of placing all the functions in a single script.

The architecture consisted of five main layers:

a. User-interface layer – collected parameters and displayed results.

b. Input layer – read schema, master-data and template files.

c. Domain layer – represented alarms, stations and related information.

d. Service layer – contained the generation rules for each test scenario.

e. Output layer – formatted and exported the generated datasets.

Separating the application into layers made it easier for me to identify where changes should be made. For example, changes to a sensor-generation rule could be made in the service layer without redesigning the whole interface.

• User-Interface Design:

The application was divided into 6 tabs:

1. Health Testing.
2. Critical Alarm Logs
3. Condition Testing
4. Condition Result
5. Health Data Extender
6. Prediction

Each tab followed a similar workflow:

Load the required reference information -> Select test parameters -> Generate data -> Preview output -> Download CSV file

This consistent workflow made the different functions easier to understand. I also provided default templates for normal use and optional upload controls for updated templates.

a. Parameter-Driven Data Generation: The application processes user-defined to generate configurable, realistic test datasets.

a. Database Integration: Generated database-ready datasets compatible with PostgreSQL and InfluxDB, enabling seamless import for software testing and system validation.

b. Modular Application Design: Consolidated multiple standalone data generation modules into a single Streamlit application, improving maintainability, usability and workflow efficiency.

 (I) Implement

The implementation phase focused on developing a modular, user-friendly application that automates the generation of realistic test datasets for the Platform Screen Door Condition-Based Monitoring (PSDCBM) system.

1. Core Application Implementation: I developed the full-stack application (py), which includes:

a. Flask Backend & Web Dashboard: I built the backend using Flask to serve a multi-page web dashboard and provide a RESTful API for all system functions. The user interface is broken down into several distinct pages for intuitive navigation and control:

i. Login & Camera Dashboard: A secure login page (html) authenticates users before granting access to the main Camera Dashboard (camera_list.html). This central hub displays all configured cameras and serves as the primary navigation point.

ii. Camera Control Page: Each camera has a dedicated control page (html) where users can view the live video feed, manually start and stop recordings, and trigger a simulated incident.

iii. Video Galleries: The system provides two distinct, sortable galleries for reviewing footage. The Recordings Gallery (html) lists all standard, continuous recordings, while the Incident Video Gallery (incident_vid.html) exclusively shows clips that have been saved as part of an incident. Both galleries support

in-browser playback and deletion.

iv. Camera & Sensor Management: I created dedicated pages for system configuration:

1. An Add/Delete Camera interface for managing the roster of cameras.
2. A Settings Page (html) allows users to define camera-specific parameters, such as video duration and the maximum number of files to keep for the automated pruning system.
3. A Sensor Information Page (html) provides a space to document physical sensor details (e.g., type, location) and view a summary of statistics and incident trigger timestamps for that camera.

2. Database Management: The system uses a robust SQLite database (db) as the single source of truth for all recorded video assets, offering a lightweight, serverless, file-based solution without a separate database server. My management strategy included several key components:

a. Dual-Table Schema for Data Segregation: I designed a specific two-table schema to logically separate routine recordings from critical incident footage.

i. video_metadata: This table logs every standard video segment created by the continuous recording

ii. incident_video_metadata: This table is exclusively for videos that have been flagged and copied during an

iii. This separation allows the automated pruning system to safely delete old, routine recordings from the video_metadata table without ever touching the permanent records in the incident_video_metadata

b. Automated Lifecycle and Storage Management: The database is the engine behind the system’s autonomous

i. Transactional Integrity: When a new video file is saved its metadata is immediately inserted into the corresponding database table.

ii. Automated Pruning: The system includes an automated pruning function that runs periodically. It queries the video_metadata table for the oldest entries for a specific camera, deletes those records from the database, and then removes the corresponding video files from the This prevents orphaned files or missing database entries.

c. Dedicated Database Management Application: As the direct administrative access to the database could be crucial, I developed a separate, standalone Flask application: py. This tool connects to the same videos.db file but serves a different purpose than the main operational server:

i. Centralized Viewing: It provides a comprehensive web interface to view, filter, and sort the entire collection of recordings and incidents from all cameras in a single

ii. Advanced Control: This utility lets an administrator directly manage any video in the database, including in-browser playback, direct download, and permanent deletion of both the record and its disk file

3. Code Management and Version Control: I used Git and GitHub to manage the codebase, commit changes with clear messages, track issues, and maintain remote backups. I refactored the project’s folders and files for better organization and updated the README This ensured code integrity and clarity while iterating on features and fixing bugs.

4. Performance Optimisation: I implemented optimisations to make the system efficient, stable, and capable of long-term autonomous

a. Reduced CPU Load and Improved Responsiveness: By offloading video encoding to an ffmpeg subprocess, I kept Python CPU usage low even with multiple cameras, preventing dashboard lag and ensuring prompt API responses.

b. Stable and Predictable Resource Consumption: I capped streams at 30 FPS and 640×480, creating a predictable performance baseline that prevented resource spikes and allowed multiple streams on low-power devices like the Jetson Nano and Raspberry Pi.

c. Faster Startup and Lower Memory Footprint: Using lazy camera initialization, I sped up application startup and kept memory usage minimal by only allocating resources when a camera was requested.

d. Elimination of Race Conditions and Enhanced Stability: I used Lock to prevent conflicts between recording and live-streaming threads, avoiding corruption and crashes common in multi-threaded camera apps.

e. Autonomous Long-Term Operation: I added an automatic cleanup feature that deletes the oldest videos when storage fills, allowing the system to run unattended for long periods.

5.  Containerization with Docker:

To solve deployment issues from complex dependencies, I used Docker to containerize the application. I wrote a Dockerfile that packages the application, its Python packages, and all system-level

dependencies into a single, portable image. This was crucial because the system relies on specific tools beyond Python, which can be difficult to configure consistently across different machines with different operating systems

a. System-Level Dependencies: The Dockerfile scripts the installation of both OpenCV and FFmpeg at the operating system level.

b. Python Environment: It uses a txt file to install the exact versions of all necessary Python packages. This creates a stable and reproducible runtime, preventing issues caused by unexpected library updates.

c.  Two key outcomes:

i. Standardized Environments: The Docker image ensured consistent behavior across all environments, eliminating the “it works on my machine” problem

ii. Simplified Deployment and Updates: It also reduced deployment to a single docker run command, making setup and updates fast, simple, and error-free.

6. Trigger Implementation: The triggers were implemented in distinct stages:

a. Manual Trigger (Deployed & Functional): This was the first trigger to be fully implemented and A button on the web dashboard successfully calls the backend API, executing the simulate_incident function as designed. This is the version currently active in the operational system.

b. Automated Triggers (Working Prototypes): To prove the feasibility of the more advanced concepts, I developed three separate, functional prototypes:

i. PLC/Hardware Trigger: A prototype script was created that successfully reads a signal from the machine’s PLC and uses it to call the incident capture function.

ii. AI Vision: Safety Compliance: A working prototype using a trained YOLOv8 model was developed. It processes the live video feed and can correctly identify a person without a helmet, triggering an incident for safety review.

iii. AI Vision: Machine Anomaly: A prototype was built to monitor a specific Region of Interest (ROI) on the machine feed. It successfully uses computer vision techniques to detect unexpected motion and trigger an

(O)  Operate

The system is currently operational in a pre-deployment testing environment.

• Current Status: The application is running as a Docker container on the Jetson Nano. All core functionalities—including live streaming, continuous recording, manual incident capture, video playback, and automated storage management—are fully functional and have been rigorously tested. The system is ready for on-site integration.
• Next Steps: The next phase of the project is to obtain a suitable casing for theJetson Nano and the power supply and physically mount onto Machine CF 653 and integrate one of the proven automated triggers (starting with the PLC trigger) for live production
• Impact: Even in its current state, the project is a successful proof-of-concept. It has validated the use of a flexible, low-cost

platform to create a powerful machine monitoring tool. It provides a solid, scalable foundation that can be easily expanded with the already-prototyped AI and PLC capabilities.

1. Contribution to Sprocket Inspection Machine

In addition to the Video Monitoring System, I had the opportunity to contribute to the Sprocket Inspection Machine, a large-scale, long-term initiative to build a new automated quality control station. This project demonstrates a different aspect of my internship experience: working as part of a larger team on a complex, multi-disciplinary engineering challenge.

Project Overview:

The Sprocket Inspection Machine is designed to be installed alongside the existing Stamping Machine STP1202 at the shop floor. Its purpose is to fully automate the quality control process. The machine’s sequence involves picking newly stamped sprockets from a conveyor using a magnetic gripper, placing them onto an SMC Slider for computer vision inspection, and then returning them to the appropriate lane on the conveyor based on the inspection result (pass/fail).

The machine is a sophisticated system composed of numerous modules, including a gantry system, a flipper mechanism, multiple pneumatic stoppers and singulators, and a central HMI for control, all integrated with the main stamping machine. My tasks were focused on three critical areas: system wiring, motion control calibration, and HMI development.

My Tasks:

1. System Wiring and Sensor Calibration: My initial task was to work on the foundational hardware and electrical systems. This involved:

a. Connectivity: Wiring and terminating essential components, including Molex connectors for power and signals, and establishing the CC-Link industrial network for communication between the PLC and various modules.

b. Sensor Calibration: I was responsible for the precise calibration of reed These magnetic sensors are critical for providing accurate position feedback from the machine’s pneumatic actuators (e.g., confirming if a gripper is fully open or closed). This work was essential for ensuring the PLC had reliable, real-time data on the machine’s state, which is a prerequisite for any automated sequence

2. Motion Control Calibration: Once the physical wiring was complete, I moved on to assisting programming the machine’s core Using the SMC ACT Controller 2 software for “teaching” and adjusting the precise positions for the two main motion systems:

a. SMC Pick & Place (PnP) Gantry: I calibrated key set points for the PnP path, including designated positions for Lane 1, Lane 2, a Reject bin, and the main Work Area where inspection

b. SMC Slider: I calibrated the slider’s positions to accommodate different sprocket sizes. This required minute adjustments to ensure that each sprocket type would be perfectly centered under the vision system.

3. Human-Machine Interface (HMI) Development: My major contribution was to assist in developing the user interface for the new machine. This was crucial for enabling testing, manual operation, and diagnostics. My work involved:

a. HMI Architecture and Navigation: I designed a menu system to organize the machine’s many functions logically.

i. Central Navigation Hub (Main Screen): I created a new “Main” screen to act as the central entry point for the Sprocket Inspection Machine.

1. Sprocket Model Selection: A quick-access panel for changing the active production model.

2. Manual Control: Navigation buttons that lead to detailed manual control pages for each major station (Slider, Flipper, Gantry, Datum, Stopper).

b. Detailed HMI Screen Development: Based on this architecture, I implemented the following key screens:

i. Production Model Selection Screen: I developed a “Model Selection” screen that chooses the specific sprocket being Selecting a model automatically loads pre-configured parameters and motion control setpoints into the PLC.

ii. System Diagnostics and History Screens:

1. Warning History: Logs all machine alarms and events with timestamps to help identify recurring
2. PLC I/O Lists (Input/Output): Shows real-time status of every digital input and output, enabling the maintenance team to quickly verify sensors and

c. System Integration: A critical final step was integrating this entire HMI module seamlessly into the main HMI of the parent Stamping Machine (STP1202). I modified the existing navigation structure to include the new Sprocket Inspection Machine section. This ensures that operators have a single, unified interface to control the entire production cell, which is essential for minimizing training time and reducing the risk of operational errors.

Through these tasks, I gained invaluable experiences in the practical integration of mechanical, electrical, and software components, contributing directly to the progress of a major factory automation project.

ii. Impact and Outcome of the Project

My work on these two projects delivered tangible value by creating a new operational tool and progressing a major factory automation initiative.

ADS Video Monitoring System

This project resulted in a fully functional, low-cost diagnostic tool ready for deployment. Its primary impact will be to increase machine uptime by providing instant video replay of fault events, eliminating guesswork for engineers. This provides a rich data source for implementing permanent, data-driven solutions.

Contributions to the Sprocket Inspection Machine

For this large-scale project, my contributions were foundational. By completing the essential wiring, sensor calibration, and motion control programming, I enabled the machine’s core physical operations. My development of integrated HMI pages also ensures it will be user-friendly for operators. This critical work helped accelerate the overall development timeline.

iii. What skills are relevant for Interns/Employees

Based on my experience in the Production Engineering department, success at Shimano requires a blend of key technical and professional skills.

Technical Skills (Hard Skills):

• Multi-disciplinary Programming: Proficiency in both high-level software (like Python with Flask for the Video Monitoring System) and low-level industrial controllers (PLC/HMI logic for the Sprocket Inspection Machine) is highly valuable.
• Software Development Lifecycle Tools: Modern engineering here requires skills beyond coding, including Git for version control, Docker for containerization, SQLite for database management, and libraries like OpenCV for building robust factory solutions.
• Industrial Networking and Connectivity: A practical understanding of how machines communicate is essential. My experience wiring Molex connectors and establishing CC-Link communication was fundamental to integrating the Sprocket Machine.
• Mechanical and Electrical Fundamentals: Foundational knowledge is necessary for any automation project, including the ability to read schematics, understand sensor functionality, and design simple mechanical components.
• Workplace Safety and Health (WSH) Certification: I also earned the WSQ certification in ‘Apply Workplace Safety and Health in

Metal Work‘, formalizing my ability to work safely in the manufacturing environment.

Professional Skills (Soft Skills):

• Adaptability: I found it crucial to be able to switch contexts between a modern software project and a traditional industrial automation
• Problem-Solving: The real value lies in identifying a problem, breaking it down, and methodically finding a This was central to debugging the Python application and calibrating the machine’s precise movements.
• Proactive Communication: Regularly updating my supervisor, asking clarifying questions, and documenting my work were essential for keeping my tasks aligned with the team’s goals.

 iv. Company Expectations

At Shimano, my role went beyond just technical tasks. I was expected to be proactive, take ownership of my projects, and seek solutions independently. High standards for quality, safety, and documentation were non-negotiable, and I learned that collaborating effectively was essential to contributing to the team’s success.

v. Any other interesting notes about the Internship

Three key observations from my internship stand out:

• The “Gemba” Philosophy in Practice: I saw engineers on the factory floor daily, ensuring solutions were practical and
• A Strategic Blend of Technology: The company pragmatically used both traditional hardware (PLCs) and modern software (Python, AI), always choosing the right tool for the job.
• Documentation as a Discipline: I learned that thorough documentation is a core part of professional engineering, ensuring systems are maintainable and knowledge is transferred.

3. About the Industry

Shimano (Singapore) is a world leader in the high-precision manufacturing of components for the cycling and fishing markets.

3.1. How the industry is Organized

3.1.1. Who are the Company’s Customers

Shimano’s Business-to-Business (B2B) model serves Original Equipment Manufacturers (OEMs) and the Aftermarket, spanning core segments like Bicycle Components, Fishing Tackle, and Rowing Equipment. The Singapore factory manufactures components for various bicycle series, including MTB, Road, City, and E-Bike ranges

3.1.2. The Industry Environment. How does the company interact with its suppliers/competitors?

The industry is competitive and technology-driven. Shimano builds

long-term partnerships with select suppliers for high-grade materials and competes with players like SRAM, Campagnolo, and Foxconn, focusing on innovation and system integration.

3.1.3. Business Model

Based on Product Leadership and Operational Excellence, Shimano ensures premium quality and OEM contracts by controlling the entire manufacturing process in-house, under a robust Quality Management System, validated by extensive ISO certifications.

3.1.4. Business Strategy

Shimano’s strategy rests on three pillars: Global Manufacturing Hub, with 48 subsidiaries across 31 countries and Singapore serving as a strategic hub overseeing 13 regional subsidiaries. Vertical Integration, controlling

in-house production for quality and innovation; and Creating Market Demand, building end-user brand loyalty.

3.1.5. Interesting Features of this industry

High Innovation: Rapid technological advancements necessitate continuous R&D.

Passion-Driven Market: Enthusiast customers and employees create a unique dynamic.

Economic Resilience: The industry benefits from global trends in health, environmentalism, and sustainable transport.

3.2. Careers in the Industry

3.2.1. Career Prospects

Singapore’s advanced manufacturing sector offers strong opportunities at Shimano in areas like Automation and System Control Engineering (My area), Mechanical Design, and Quality Assurance. Skills gained are highly transferable to sectors such as aerospace and electrical engineering.

3.2.2. Recruitment and Career Progression

Shimano hires engineering graduates from polytechnics and universities, with internships as a key talent pipeline. Career growth follows the Kaizen principle, with high performers advancing to senior technical or management roles, potentially including overseas assignments.

3.2.3. Training of new employees

New hires and interns receive structured on-the-job training and mentorship covering Workplace Safety and Health (WSH), SOPs, and company practices. WSH training aligns with Shimano’s yearly objectives to reduce hand injuries. Continuous learning is supported through various Training Leave options.

3.2.4. Attractiveness of the Company, and the Industry Sector. Employee Retention

Shimano appeals as an employer through its global brand, advanced automation, and Singapore’s manufacturing stability. Retention is strengthened by a collaborative “Team Shimano” culture, pride in its products, and strong investment in employee growth.

3.2.5. Interesting notes about Careers in this Industry Sector

This sector offers the satisfaction of seeing products you helped create used worldwide, aligning professional engineering with personal passions like cycling or fishing, while providing cross-cultural experience in a Japanese multinational.

4. Reflection and Learning Points

4.1. Reflection on CDIO Skills

4.1.1. How the CDIO skills learnt in SP helps you in your work

The CDIO framework provided a crucial structure for my projects. It guided my hardware evaluation during the ‘Conceive‘ phase and prompted me to use project management tools like Git during the ‘Design‘ phase. This ensured the system I built would be maintainable, for Implementation and Operation during the predeployment phase.

4.1.2. What are the CDIO skills that you had utilised in getting the work done

I utilized System Thinking to design the ADS monitoring ecosystem, from PLC triggers to the user dashboard. I also applied iterative testing and refinement to calibrate the precise movements of the Sprocket Inspection

Machine, ensuring its mechanical accuracy through repeated, incremental adjustments.

4.2. Recommendations for the Company

4.2.1. Learning Points from your attachment that benefits the company

My work on both a software-driven system and a traditional machine highlighted a key opportunity. I would recommend Shimano focus on creating a standardized software layer to connect its reliable

PLC-controlled machines with modern data analysis and AI tools, unlocking new insights without replacing proven hardware.

4.2.2. How has your work contributed to the company?

My work delivered a functional diagnostic tool (ADS System) that will reduce machine downtime. I also accelerated the development of a major automation initiative (Sprocket Inspection Machine) by completing its foundational motion programming and system wiring, bringing it closer to full operation.

4.2.3. Anything else that the Company does to remain competitive?

From my observations, Shimano’s strength lies in its blend of traditional and modern technology. I believe its competitiveness is reinforced by a commitment to continual improvement, embedded within Integrated Management System (IMS) and supported by extensive ISO certifications. This drives systematic enhancements across quality, environment, health, safety and energy performance.

4.3. Careers in the Industry

4.3.1. My own lessons learnt

I learned three key lessons:

• Technical: The importance of building robust, maintainable software that meets industrial standards, beyond just academic
• Professional: The value of taking ownership, and ensuring work is thoroughly documented and clearly communicated to the team.
• Personal: How a shared culture, like “Team Shimano,” can inspire motivation and uphold high quality

4.3.2. Suggestions and recommendations to better prepare students for the workplace

To better prepare students, I recommend three actions:

• Mandate the use of professional tools like Git and Docker in all project coursework.
• Increase the number of academic projects that require

hardware-software integration.

• Introduce realistic project constraints like budgets and legacy system requirements to foster a business-aware engineering

4.4 Conclusion

My internship at Shimano (Singapore) has been an invaluable experience, providing me with practical insights into the world of industrial automation and software development. The opportunity to work on both the Video Monitoring System and the Sprocket Inspection Machine has allowed me to apply my academic knowledge to real-world challenges and develop a deeper understanding of the entire engineering lifecycle, from conception to operation. I am grateful for the mentorship and guidance I received from my supervisor and the System Control team, and I am confident that the skills and lessons learned will be highly beneficial in my future career.