ABSTRACT Mathematical problem-solving is an essential competency that needs to be developed from elementary school; however, its development remains challenging, particularly in learning whole-number concepts. This study aimed to analyze the effects of the Problem Based Learning (PBL) model and learning discipline on students’ mathematical problem-solving ability, both partially and simultaneously. A quantitative approach with a quasi-experimental nonequivalent control group design was employed. The sample consisted of 68 fifth-grade students, comprising 38 students from SDN 7 Belinyu as the experimental group and 30 students from SDN 5 Belinyu as the control group. Data were collected through written pretests and posttests and a learning discipline questionnaire, then analyzed using the independent-samples t-test, Analysis of Covariance (ANCOVA), multiple linear regression, and N-Gain, with the assistance of SPSS 25. The results showed that the experimental group achieved a mean posttest score of 79.18, higher than the control group’s score of 63.63. This difference remained statistically significant after controlling for initial ability. Learning discipline was also positively and significantly associated with problem-solving ability. The regression model explained 60.8% of the variance in this ability. The experimental group’s N-Gain was 0.7621, categorized as high, whereas the control group’s N-Gain was 0.5872, categorized as moderate. These findings support the use of PBL as an alternative approach to mathematics instruction focused on problem-solving, while taking learning discipline and the limitations of the research design into account. ABSTRAK Kemampuan pemecahan masalah matematika merupakan kompetensi penting yang perlu dikembangkan sejak sekolah dasar, tetapi pembelajarannya masih menghadapi tantangan, terutama pada materi bilangan cacah. Penelitian ini bertujuan menganalisis pengaruh model Problem Based Learning (PBL) dan kedisiplinan belajar terhadap kemampuan pemecahan masalah matematika siswa, baik secara parsial maupun simultan. Penelitian menggunakan pendekatan kuantitatif dengan desain quasi-experimental berbentuk nonequivalent control group design. Sampel terdiri atas 68 siswa kelas V, yaitu 38 siswa SDN 7 Belinyu sebagai kelompok eksperimen dan 30 siswa SDN 5 Belinyu sebagai kelompok kontrol. Data diperoleh melalui tes uraian pretest–posttest dan angket kedisiplinan belajar, kemudian dianalisis menggunakan uji independent-samples t-test, Analysis of Covariance (ANCOVA), regresi linear berganda, dan N-Gain dengan bantuan SPSS 25. Hasil menunjukkan bahwa rata-rata posttest kelompok eksperimen mencapai 79,18, lebih tinggi daripada kelompok kontrol sebesar 63,63. Perbedaan tersebut tetap signifikan setelah kemampuan awal dikendalikan. Kedisiplinan belajar juga berhubungan positif dan signifikan dengan kemampuan pemecahan masalah. Model regresi menjelaskan 60,8% variasi kemampuan tersebut. Nilai N-Gain kelompok eksperimen sebesar 0,7621 berkategori tinggi, sedangkan kelompok kontrol sebesar 0,5872 berkategori sedang. Temuan ini mendukung penggunaan PBL sebagai alternatif pembelajaran matematika yang berorientasi pada pemecahan masalah, dengan tetap memperhatikan kedisiplinan belajar dan keterbatasan desain penelitian.