The Army’s transformation mandate is unambiguous: deliver warfighting capability 25–30% faster. Every Tier 2 metric published in support of that mandate – days between milestones, days to complete the requirements process, days to complete contracting, days to complete testing – is really a decision throughput measurement. Yet the systems engineering (SE) discipline that governs those timelines has no formal production framework for producing decisions. This paper proposes Decision Engineering as the framework. Grounded in lean production theory, applied to information work and anchored to the defense acquisition policy structure of DoDD 5000.01 and DoDI 5000.02, Decision Engineering reconceives SE as the discipline of designing, operating, and continuously improving the lifecycle decision production system. It introduces a formal decision ontology comprising six decision states (Latent, Declared, Active, Deferred, Closed, Reopened), four topology relationship types (Precedes, Enables, Constrains, Triggers), and five diagnostic biomarkers that measure production system health. The paper presents the Decision Factory model and describes how its application to ground vehicle system programs and across the Army acquisition enterprise can accelerate capability delivery without sacrificing rigor. Defense acquisition programs do not fail for lack of engineering rigor or digital tooling. They fail because the production system that converts information into decisions is ungoverned – no inventory count, no flow discipline, no throughput measurement. Decision Engineering names this production system, provides instruments to measure its health, and offers a practitioner framework for running it better.