Ezekiel L. Bernardo

The adoption of Generative Artificial Intelligence (GenAI) is an engineering systems problem. It is a complex sociotechnical system governed by feedback loops, time delays, and nonlinear interactions that resist conventional management and cannot be analyzed with static methods. Despite rapid uptake, enterprise GenAI adoption has plateaued below 50% of potential users, following a rise–dip–rebound–plateau trajectory that cross-sectional adoption models cannot reproduce. This study develops a System Dynamics simulation model as a decision-support tool for governing GenAI adoption. The model endogenizes technical performance (accuracy, output similarity, hallucinations), the developer and model supply ecosystem, user literacy, trust and trauma dynamics, costs, and interventions, with the adopter population segmented across Rogers' five diffusion categories. Integrating 17 reinforcing and 8 balancing feedback loops, it is validated against an empirical reference mode through behavior-pattern and structure-oriented tests, reproducing the observed trajectory as an endogenous outcome. Sensitivity analysis identifies five leverage points, dominated by developer-ecosystem structure and risk exposure rather than cost or access, and yields four engineering findings: (1) developer-ecosystem growth beyond governance capacity reduces long-run adoption (the AI sprawl effect); (2) trust in AI outputs decays rapidly and recovers asymmetrically; (3) raising user literacy without commensurate reliability improvement reduces adoption among later segments; and (4) interventions interact nonlinearly, with combined deployment producing outcomes inferior to targeted single interventions. Three intervention strategies are designed from the leverage points and evaluated, providing a reusable, quantitative decision-support framework for engineering the responsible deployment of GenAI systems.

With an estimated 1.8 billion individuals menstruating globally, menstrual product selection plays a critical role in health, comfort, and overall well-being. This study aims to address a gap in menstrual health and hygiene by examining ergonomic considerations in the redesign of a sustainable menstrual product. A comprehensive needs assessment was conducted through focus group discussions and online surveys to identify user preferences and challenges. The design methodology incorporated quality function deployment (QFD) and design failure mode and effects analysis (DFMEA), followed by CAD-based prototyping and expert evaluation. Findings supported a tampon-shaped menstrual cup design, integrating ease of use with the environmental benefits of menstrual cups.

Ensuring accessibility for wheelchair users is essential in promoting inclusivity and equal opportunities, yet challenges remain, especially regarding infrastructure and public furniture. One common issue is the incompatibility of standard public tables with wheelchair heights, which often results in discomfort or discourages use. While various wheelchair trays have been introduced, many lack stability, ergonomic design, and ease of attachment. To address these shortcomings, we developed AdjusTABLE—a portable, height-adjustable, and foldable tray designed with ergonomic comfort and user convenience. Using empathy maps and focus group discussions, we created a tray that holds drinks, pens, phones, and personal belongings, improving usability and independence.

User interfaces (UIs) have become prevalent and dominant to maximize overall user experience. In order to enhance UI design, the effects of their base properties must be mapped out. Therefore, we constructed a structural equation model, considering the direct effects of design properties and the mediating effects of user performance, with the specific goal of maximizing the UI for users. A total of thirty-six participants were surveyed with different random combinations of UIs while using an eye tracker and measuring their final perception. Significant direct, mediating, and moderating effects were found in this study. Design guidelines were made for maximizing user experience in this study.

Managing electronic waste (e-waste) in the Philippines is a critical challenge, no with roughly 80% handled by an informal sector using hazardous methods. This study develops a context-specific Circular Economy (CE) framework for urban Manila by quantifying the behavioral, institutional, and socio-economic factors influencing recycling efficiency. Using a hybrid methodology, quantitative data were collected from 435 informal recyclers. Structural Equation Modeling (SEM) supported 16 of 18 hypothesized pathways from the Theory of Planned Behavior (TPB), though Perceived Behavioral Control did not directly affect Intention. An Artificial Neural Network (ANN) sensitivity analysis identified economic factors, Income Level (84.01%) and Financial Incentives (82.86%), as the dominant predictors of behavior, followed by the Cultural–Cognitive Pillar (80.98%). This necessitates modifying the TPB for subsistence economies, where economic survival acts as a super-moderator. The resulting CE framework mandates inclusive policies, prioritizing “Economic First Interventions” like buy-back schemes to equitably integrate informal recyclers into formal Extended Producer Responsibility systems.