Valuing supply‐chain responsiveness under demand jumps

Abstract

As the time between the decision about what to produce and the moment when demand is observed (the decision lead time) increases, the demand forecast becomes more uncertain. Uncertainty can increase gradually in decision lead time, or can increase as a dramatic change in median demand. Whether the forecast evolves gradually or in jumps has important implications for the value of responsiveness, which we model as the cost premium worth paying to reduce the decision lead time (the justified cost premium). Demand uncertainty arising from jumps rather than from constant volatility increases the justified cost premium when an average jump increases median demand, but decreases the justified cost premium when an average jump decreases median demand. We fit our model to two data sets, first publicly available demand data from Reebok, then point‐of‐sale data from a supermarket chain. Finally, we present two special cases of the model, one covering a sudden loss of demand, and the other a one‐time adjustment to median demand.

Inventory agility upon demand shocks: Empirical evidence from the financial crisis

“…This reasoning, however, does not factor in credit and financial limitations that can, in such situations, steer inventory investment decisions. While there is recent evidence suggesting that demand jumps have an impact on the break even point of investments geared toward reducing lead‐times (Bicer et al, 2018), to the best of our knowledge, no study explicitly addresses a manufacturer's inventory management capabilities in times of crises or the causes and effects of the resulting inventory dynamics. The contribution most related to our topic is a recent paper by Kesavan and Kushwaha (2014), who study retail inventory investment during macroeconomic expansions and contractions.…”

Editorial: New developments at the Journal of Operations Management

“…Another type of EGA paper explores whether the extra work and complexity involved in improving the accuracy of a model is warranted by the improvement in decision making. For example, Biçer et al (2018) investigated whether a model that differentiates between usual demand volatility and shifts in median demand provides enough extra guidance to warrant the extra complexity or if it is sufficient to approximate demand uncertainties with a model that assumes usual demand. Biçer et al (2018) showed that it depends on the expected direction of the shift in median demand: For a shift that is expected to be positive, the implications for the value of responsiveness suffice to justify use of the more complex model.…”

“…For example, Biçer et al (2018) investigated whether a model that differentiates between usual demand volatility and shifts in median demand provides enough extra guidance to warrant the extra complexity or if it is sufficient to approximate demand uncertainties with a model that assumes usual demand. Biçer et al (2018) showed that it depends on the expected direction of the shift in median demand: For a shift that is expected to be positive, the implications for the value of responsiveness suffice to justify use of the more complex model. When, however, median demand is expected to shift down, assuming that all demand uncertainty comes from usual demand volatility provides a reasonable approximation.…”

Beyond Retail Stores: Managing Product Proliferation along the Supply Chain

“…We model the evolution of demand forecasts D i from t 0 to to t n according to the multiplicative martingale model (m-MMFE), which is known to fit very well to empirical data of demand-forecast updates (Heath and Jackson 1994, Oh and Özer 2013, Biçer et al 2018. Let (Ω, F, P) to denote a filtered probability space on which demand forecasts follow the m-MMFE process.…”