In a research study, collected and processed data are needed to solve problems and prove hypotheses. However, datasets often contain missing or null values. One way to overcome the missing values problem is by using imputation techniques. The technique works by filling in the missing values with an estimated weight that has been analyzed to create a complete dataset. In the process, researchers usually found the data used for imputation to have unclear or inconsistent characteristics that may lead to bias. This issue can be addressed by implementing the Fuzzy C-Means (FCM) method to estimate the missing values and improve the data quality. However, estimating missing values using the FCM model produces predictive models with various parameters; hence, another approach to creating the best model with optimal parameters. Therefore, this underlies the need for an ensemble system combining different machine learning models to earn the best estimation result of missing values, including the fuzzy machine learning models. The ensemble system in this study uses majority voting and averaging, which can help boost accuracy without making the FCM system more complex. This research paper is born of the novelty of the combination of both designs through Ensemble FCM Model with Majority Voting and Averaging research topic. This topic in this study works to impute the missing values of Chronic Obstructive Pulmonary Disease (COPD) data in 2012-2017 from Cipto Mangunkusumo Hospital (RSCM) established the actual data. In addition, this study can help the hospital predict the exacerbation of COPD patients in the future. The random forest classification is used to create a prediction more trusted. As a result, this research paper compares the FCM model with and without the ensemble to prove the performance improvement. © 2024 American Institute of Physics Inc.. All rights reserved.