We first propose a Geo-Data Fusion Integrator. Specifically, we design a sequential-parallel-modularized (SPM) approach to integrate different datasets into a geo-data object, i.e., a multidimensional unified-OLAP cube, archived in a geo-data warehouse for decision-making analysis. Different datasets of geo-data objects are processed in parallel across multi-stages in sequence, and then integrated into a well-defined OLAP cube. Each SPM component is a self-contained, modularized unit that processes the data. The technical merits of this SPM approach include fast manipulations, error minimization, and easy maintenance. Second, to create a unified geo-data object, we extend the object-oriented spatial-temporal data model as a multidimensional OLAP cube, i.e., a Star-based Geo-Object-Oriented SpatiotEmporal (S-GOOSE) data model, which combines the advantages of both OLTP and OLAP approaches. This S-GOOSE data model is an object-relational-based cube that enables military operators to analyze unified geo-data objects from multiple dimensions, such as time, space, and location, to help them make a better decision on paths.