Mikania micrantha, one of the world’s most invasive alien plant species, poses a severe threat to agriculture and ecosystems in Taiwan. Conventional control methods, such as manual removal, chemical herbicides, and biological agents, face limitations including high labor demand, environmental risks, and uncertain effectiveness. This study aims to design and evaluate an IoT-based smart weed control system that enhances control efficiency while reducing labor and environmental impacts. The research follows the Design Science Research (DSR) methodology to iteratively develop the system through conceptual design, prototype development, and field testing. A Control Efficiency Index (CEI) was established to quantitatively assess weed suppression performance. In addition, user feedback was collected using the Technology Acceptance Model (TAM), and a Cost-Benefit Analysis (CBA) was conducted to evaluate economic feasibility. Field experiments in orchards in Miaoli, Taiwan, demonstrated that the system operated reliably, achieving a CEI 87% and reducing manual labor costs by approximately 33%. The TAM survey indicated high user acceptance, with perceived usefulness and behavioral intention both exceeding 4.0 on a five-point Likert scale. The CBA results revealed a payback period of about 1.67 years, with the annual benefit-cost ratio (BCR) rising to 1.6 from the second year onward, highlighting the system’s long-term economic value. Overall, the proposed IoT-based weed control system effectively mitigates the spread of M. micrantha, reduces reliance on manual labor and chemical herbicides, and demonstrates practical and economic viability. This research not only provides a novel solution for invasive species management but also contributes empirical evidence to the advancement of sustainable smart agriculture.