Mathematical problem-solving ability is an important ability for students to master, but it is often not facilitated due to the lack of use of adequate teaching materials and learning models. This study aims to examine the effectiveness of SMART-KU local wisdom-based module in Problem-Based Learning on students' mathematical problem-solving ability. Using a quantitative method of quasi-experimental design, which specifically takes a non-equivalent control group design, with a total sample of 32 fifth-grade students of SDN 1 Mijen in the experimental class and control class, and data analysis through paired sample t-test, independent sample t-test, and N-Gain score test. The results showed that: (1) through paired sample t-test, there is a significant average difference in students' mathematical problem solving ability before and after treatment, both in experimental (Sig. (2-tailed) = 0.000 (≤ 0.05)) and control classes (Sig. (2-tailed) = 0.000 (≤ 0.05)), (2) through independent t-test, there is a significant average difference in students' mathematical problem solving ability between experimental and control classes (Sig. (2-tailed) = 0.000 (≤ 0.05)), and (3) through N-Gain test, students' mathematical problem solving ability in experimental classes (0.87) improved better than students in control classes (0.68). It can be concluded that the SMART-KU module based on local wisdom in PBL effectively improves students' mathematical problem-solving ability. This research contributes to alternative learning to improve students' mathematical problem-solving ability, which combines local wisdom-based teaching materials and contextual and meaningful learning models.