In the context of small and medium-sized enterprises (SMEs) in the Peruvian garment sector, where production efficiency typically ranges between 55% and 65%, this study proposes an integrated model aimed at reducing operational losses and optimizing system performance. The solution is built on three core pillars: work standardization aligned with Industry 4.0 principles, predictive and planned maintenance using convolutional neural networks (CNN), and operations scheduling supported by artificial intelligence (AI) algorithms. The model was implemented in a Peruvian garment company facing high operational variability, frequent human errors, and weak planning capacity, resulting in annual losses equivalent to 9% of its total revenue. The implementation led to a reduction in non-productive time, improved machine availability, and enhanced adherence to the production schedule, all achieved with a low-cost investment. The outcomes were validated through key performance indicators (KPIs), showing significant improvements: annual lost hours reduced from 344 to 300, machine availability increased from 65.80% to 85%, and postponed orders decreased from 20% to 10%.