We introduce a new three-parameter model called the logistic Nadarajah–Haghighi distribution. An important property of the proposed distribution is that it can produce decreasing and unimodal failure rates. These behaviors can arise in several reliability problems. We provide some useful quantile measures and other reliability properties. We present the maximum likelihood estimators of the model parameters. We perform applications that show the usefulness of the new distribution for modeling mechanical components failure times and impact factors in the informetrics literature. Further, we present a parametric regression model for right-censored data considering a median-based parametrization of the new distribution. A Monte Carlo study is carried out to evaluate the performance of the estimators. Finally, we analyze the reliability of pressure vessels wrapped with a Kevlar fibre using the proposed regression model.