Demand-Driven Material Requirements Planning (DDMRP) is a hybrid push–pull strategy aimed at buffering variability
and enhancing flow. The performance of DDMRP is determined by two interrelated choices: (i) the parameterization of
buffer zones (e.g., variability and lead-time factors, MOQ, thresholds); and (ii) the strategic positioning of decoupling
points. Traditionally, planners relied on their judgment instead of systematic approaches. This led to a divergence
between theoretical potential and achievement. This paper presents a systematic review of the publication period from
2020 to 2025 within the context of transforming DDMRP via heuristics into optimization-based formulations. Two
principal streams emerge. The parameter optimization models ranging from linear mixed integer programming to multi
objective metaheuristics and fuzzy/statistical based designs consistently offer improvements over heuristics in reducing
inventories and shortages while increasing service levels. The buffer positioning models focus on MILP/MINLP,
graph-theoretic explorations, and hybrid heuristics to formalize decoupling decisions; this stream is less evolved and is
typically not coupled with parameter tuning, with few studies addressing their integration, creating the largest gap. The
findings from the review point to three principal conclusions: (1) optimization consistently outperforms heuristics; (2)
achieving ideal service targets (100% on-time delivery) requires exponentially high inventories or capacities; and (3)
parameter optimization appears to be more advanced than buffer positioning, which remains discrete and deterministic.
We conclude with a future research agenda focused on the unifying and dynamic co-optimization of buffer positioning
and parameters, considering uncertainty and capacity constraints. Thus, linking the models through a rolling horizon
and adaptive methodology would be necessary to advance DDMRP from heuristics to optimization science.

Keywords

Demand Driven MRP, DDMRP, Decoupling Points, Customer Order Decoupling Point, Inventory buffer