The application of point process models to the natural sciences and engineering have grown significantly in the past half century. While techniques for modeling and estimation are well established in the literature, the application of residual analysis techniques for model evaluation is still infrequent in the literature, especially for multi-dimensional models. This dissertation brings modern point process methodology to the area of wildfire hazard assessment. We use space-time point process models and apply residual analysis methods to evaluate a wildfire hazard index using extensive data from Los Angeles County, California. In addition we develop non-parametric methods for estimating the renewal distribution of a space-time coverage process. Finally, a software package ptproc is developed for analyzing multi-dimensional point process models.