RESEARCHERS from Ateneo de Manila University have developed an artificial intelligence (AI) system that helps small food stalls and convenience stores digitize handwritten logbooks and analyze inventory data to improve operations.

The system, developed by the university’s Business Insights Laboratory for Development (BUILD), uses AI to scan logbook photos, recognize products, match prices and generate sales summaries.

“The system allows even someone with no digital training to grasp inventory trends with ease,” BUILD said in a statement on Monday. “This helps businesses quickly identify bestsellers or slow-moving stock, thereby making it easier to keep up with customer demand.”

Small neighborhood stores and food stalls in the Philippines have traditionally relied on pen-and-paper ledgers to manage sales, inventory and deliveries. This analog system limits data analysis and slows business decisions, according to the researchers.

The team introduced their AI “copilot” model after observing hesitation among many micro, small and medium enterprises in adopting tech due to “steep learning curves and job redundancy.”

“AI lends itself perfectly to business data analysis: it makes short work of identifying which products are performing well or poorly; tracking sales trends over time; and offering recommendations on inventory, pricing, and restocking,” they said.

The prototype system runs on Python, using Amazon Web Services for optical character recognition and Anthropic’s Claude 3 Haiku large language model to interpret handwritten logbook entries.

Initial testing was conducted at a student-run food stall in Ateneo’s enterprise center. Researchers said the system achieved moderate accuracy and showed potential for broader applications.

The team noted that the tool could be adapted to process other types of handwritten records including inventory sheets, delivery logs and payroll ledgers.

“Not unlike analog logbooks themselves, this AI tool is meant to be simple, affordable and easily upgradable,” the researchers said. “As AI accuracy improves by training on more shorthand writing, local stalls can eventually gain reliable, low-cost access to business insights once reserved for larger enterprises.”

The initiative aligns with state efforts to modernize microbusinesses. Last year, the Department of Science and Technology supported local startup Packworks in helping mom-and-pop stores nationwide adopt AI-driven inventory systems.

The Department of Trade and Industry in June said more than 1.3 million such stores across the Philippines continue to operate with minimal capital and limited access to digital tools. — Katherine K. Chan

RELATED ARTICLESMORE FROM AUTHOR