The Department of Science and Technology (DOST) funded system to automate manual evaluation of repossessed motorcycles was officially launched on June 18, 2024, at the Asian Institute of Management (AIM), Makati City.
ARMAS is a result of an R&D project between Unistar Credit and Finance Corporation and the Asian Institute of Management funded under the DOST Collaborative Research and Development to Leverage Philippine Economy (CRADLE) Program.
The DOST CRADLE Program aims to enable technological advancement and innovation of local companies and to support the growth of the Philippine Innovation ecosystem. Through the program, the academe and/or R&D Institution in partnership with a Filipino company, undertakes R&D to improve the company’s products, processes, and services to become more competitive in their respective industries. Beneficiary-companies under this program must provide at least 20% counterpart funding and commit to adopt the technology that will be developed. This way, research outputs from the academe are directly translated commercially to the market.
In her keynote speech, DOST Undersecretary for R&D Leah J. Buendia said: “The benefit of this collaboration enabled Unistar to potentially lower their operational costs and increase their revenue. Furthermore, since this is a pioneering work for this sector, it proves that AI-driven strategies can boost the competitiveness of any business. The success of ARMAS is a testament to the effectiveness of the DOST CRADLE Program.”
: Dr. Enrico C. Paringit, Executive Director of DOST PCIEERD, during his welcome address at the PROJECT ARMAS launch. He reaffirmed the innovative work of AIM and UNISTAR in addressing the necessity to emphasize the usefulness of AI in evaluating motor vehicle conditions quickly and accurately.
Dr. Ericka Fille T. Legarda of the Asian Institute of Management and the project leader for the Unistar Automated Repossessed Motorcycle Assessment System (Unistar ARMAS) during her discussion on the success of the initiative.
The project team of AIM with the officers of UNISTAR together with Dr. Enrico C. Paringit, Executive Director of DOST PCIEERD (fifth from right), and other officials and the project team from PCIEERD.
Photo of the ARMAS application on mobile phone
Photo of the actual demonstration on how the system works
The Unistar ARMAS project developed an AI-based decision support tool for streamlining asset recovery operations of Unistar Credit and Finance Corp, a local credit and financing institution for the motorcycle market. ARMAS uses artificial intelligence and machine learning, including computer vision models for image segmentation and object detection, as well as deep learning–based sound analysis, leveraging on Unistar's unique and rich datasets.
The system entails the creation of an automated pipeline for data preprocessing, which includes ingestion, cleansing, and harmonization, to render the disparate data amenable as inputs for the AI models; this pipeline will help speed up operations by reducing the amount of time needed to perform preprocessing, which is critical for AI model performance.
ARMAS will make Unistar’s internal operations more efficient, enabling the company to provide faster and better services, by reducing the number of manhours spent on manual processes.
“This project represents a pioneering effort in the local industry, addressing the need to highlight the value of AI in assessing vehicles conditions in a rapid yet reliable manner. AI-driven strategies like Unistar ARMAS are crucial for bolstering the competitiveness of businesses in today’s fast-evolving market,” DOST Philippine Council for Industry, Energy and Emerging Technology Research and Development Executive Director Dr. Enrico C. Paringit said.
