Java has become one of the most popular programming languages for application project development nowadays, due to its rich dependency libraries and convenient build tools such as Maven and Gradle. However, with the continuous increase in the scale of dependency libraries, the dependency management of Java projects becomes increasingly complex and constantly exceeds the management capabilities of existing tools. The potential problems are likely to be triggered unexpectedly, seriously affecting the building and running of the current project and other projects in the Java ecosystem, such as causing build errors, runtime crashes, or semantic conflicts. This study aims to address the gaps in the analysis of dependency management issues found in existing research and technical literature by introducing the concept of “dependency smell”, to build a unified model for these challenges. This study conducts a comprehensive empirical study on dependency management issues, covering all categories of Maven and Gradle related problems. This study analyzes diverse dependency management issues gathered from open-source communities (e.g., GitHub), official documentation (e.g., Maven manual), as well as various surveys and technical papers. Finally, 13 types of dependency smell, as well as their triggering roots and impact characteristics, are summarized. Based on the findings of this empirical study, a unified detection algorithm for dependency smells in Java projects is designed, and a special detection tool JDepAna suitable for Maven and Gradle build tools is implemented. Experimental results demonstrate that for known dependency smells, JDepAna achieves a detection recall rate of 95.9%. For hundreds of new Java projects, JDepAna detects 30689 instances of dependency smells. 360 instances are selected, and the true positive rate of manual verification reaches 96.1%. Additionally, this study reports 48 instances to developers, with 42 instances promptly confirmed and 21 promptly fixed, thereby validating the efficacy and practicality of the proposed Java dependency smell detection algorithm and tool in facilitating quality assurance for Java projects.