Maximiliano Udenio

The pork cycle exhibits persistent oscillations in prices and production with a wavelength of three to six years. We develop a system dynamics model in which cyclical behaviour arises endogenously from biological delays and a simple adjustment rule, without the explicit modelling of price or forward-looking optimization from decision-makers. The model reproduces the cyclical component of Dutch pork prices, matching its wavelength and turning points, and explains the gradual shortening of the cycle through technological changes that reduced production lead times. Sensitivity analysis further separates the parameters that govern the cycle’s wavelength from those that govern its amplitude, and identifies policy levers acting on each. Our findings do not imply that prices are irrelevant; rather, they highlight that a significant part of the pork cycle is structural, with prices primarily reflecting underlying production delays. This perspective complements rational-expectations models and clarifies why similar ‘pork cycle’ patterns appear in other industries with long lead times.

We study non-stationary single-item, periodic-review inventory control problems in which the demand distribution is unknown and may change over time. We analyze how demand non-stationarity affects learning performance across inventory models, including systems with demand backlogging or lost-sales, both with and without lead times. For each setting, we propose an adaptive online algorithm that optimizes over the class of base-stock policies and establish performance guarantees in terms of dynamic regret relative to the optimal base-stock policy at each time step. Our results reveal a sharp separation across inventory models. In backlogging systems and lost-sales models with zero lead time, we show that it is possible to adapt to demand changes without incurring additional performance loss in stationary environments, even without prior knowledge of the demand distributions or the number of demand shifts. In contrast, for lost-sales systems with positive lead times, we establish weaker guarantees that reflect fundamental limitations imposed by delayed replenishment in combination with censored feedback. Our algorithms leverage the convexity and one-sided feedback structure of inventory costs to enable counterfactual policy evaluation despite demand censoring. We complement the theoretical analysis with simulation results showing that our methods significantly outperform existing benchmarks.

Many countries have implemented various policies to reduce carbon emissions. Among the policies available, capand-trade regulations have well been accepted as an effective market-based mechanism to curb carbon emissions. This policy offers the opportunity for firms to share the benefits of low-emission in a carbon trading market by selling and purchasing credits. In this paper, we focus on the cap-and-trade emission reduction policy. Specifically, we study an emission-dependent supply chain consisting of a manufacturer, a supplier and a retailer. In this supply chain, a cap on emission permits for the manufacturer is primarily imposed by the government and additional emissions must be purchased from the emission trading market if the permit needed is larger than the cap. We investigate the effects of the emission cap-and-trade on the supply chain members' decisions for decentralized and centralized supply chain using a Stackelberg game model. In the decentralized supply chain, a product's retail price is higher than in the centralized decision-making mode. Moreover, the price of extra emissions under the decentralized supply chain decrease as the carbon emissions cap increases. Our results show that in the centralized supply chain enterprises obtain higher benefits than in the decentralized supply chain. The outcome of this research can help i) policy-makers to make effective emission reduction policies and ii) manufacturers and retailers to make sustainable operations and supply chain decisions.

In this paper, we study the influence of seasonal demands and forecasts on the performance of an Automatic Pipeline, Variable Inventory, Order-Based, Production Control System (APVIOBPCS) using linear control theory. In particular, we consider a system that uses a seasonal forecast based on a no-trend, additive-seasonality exponential-smoothing model, and compare its performance to an equivalent system using simple exponential smoothing. We find that the system with seasonal forecasting significantly outperforms the system with simple exponential smoothing under certain demand assumptions. With optimal parameter settings, the forecast error of the seasonal model can be up to 40% lower. However, we also find that the forecast superiority does not necessarily translate to the performance of the system measured through the bullwhip metrics. In addition, the seasonal forecasting model is very sensitive to the demand frequency and smoothing parameters, while the simple exponential smoothing model is very robust. This implies that the real life benefits of implementing a seasonal forecasting model are not obvious and depend on the particular situation; under a large number of settings (e.g. low seasonality, high-smoothing), the good performance of simple exponential smoothing certainly justifies its popularity in the industry and research worlds alike.

The objective of this paper is to identify antecedents of inventory agility (i.e., the capability to quickly adapt inventories to changes in demand) upon demand shocks based on the awareness‐motivation‐capability (AMC) framework and to explore the link between inventory agility and financial performance. We introduce an empirical measure of inventory agility based on the deviation of relative inventories (i.e., inventory days) from their forecasted values. We hypothesize that firms with higher awareness, motivation, and capabilities are associated with higher inventory agility in the presence of demand shocks. We define two empirical measures for each of the three dimensions of the AMC framework in the context of inventory agility: awareness (i.e., market orientation and technology orientation), motivation (i.e., gross margin and liquidity), and capabilities (i.e., inventory management capability and resource availability). In addition, we incorporate the constraining factor model (CFM) into the AMC framework, thus allowing for complementarity among the different measures. In this view, the influence of each of the measures on inventory agility varies according to which of the measures is the constraining factor for a given firm. The 2008 financial crisis may have tested firms' inventory agility more than any other crisis since the Great Depression, as an unprecedented collapse of demand coincided with a reduction in credit availability. Therefore, for our analysis, we use firm‐level empirical data from 1263 public U.S. manufacturing firms for the 2005–2011 period. We find that firms' motivation and capabilities are key factors associated with inventory agility. Through the CFM, we show that identifying the constraining factors leads to a more refined understanding of the moderating effects of the antecedents of inventory agility. In a separate analysis, we find that inventory agility is positively associated with a number of financial performance metrics during crisis periods. We distinguish between inventory underages and overages and find that, during the crisis, they are both associated with lower financial performance. Furthermore, we find evidence that higher underages (overages) magnify the effect of overages (underages). Among other managerial insights, our findings suggest that the use of inventory reductions as a quick way to increase liquidity must be gauged against their potential impact on other aspects of financial performance.

In this paper, we study the performance of an Automatic Pipeline, Variable Inventory, Order-Based Production Control System (APVIOBPCS) using linear control theory. In particular, we consider a system with independent adjustments for the inventory and pipeline feedback loops and the use of triple exponential smoothing (the Holt-Winters no-trend, additive seasonality model) as a forecasting strategy. To quantify the performance of the system, we derive the transfer functions of the system and plot the frequency response of the system under a number of different parametrizations. We find that the system using Holt-Winters forecasting (the HW-model) significantly outperforms the system using simple exponential smoothing (the SES model), commonly found in the literature, under certain demand assumptions. However, we find that the HW-model is very sensitive to the demand frequency, while the SES is very robust. Thus, the performance range is substantially narrower for the SES model. Finally, we show that previous insights related to behavioral biases are not affected by the choice of forecasting strategy.