A Bayesian Decision Model with Hurricane Forecast Updates for Emergency Supplies Inventory Management

Taskin, Selda; Lodree, Emmett J.

doi:10.1057/9781137535696_13

Cited by 35 publications

(8 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our study can help retailers identify consumer stockpiling propensity so that they can better predict demand and arrange to stock this essential product, given the projected path of a hurricane. Therefore, our study primarily relates to existing research on inventory pre‐positioning in the path of hurricanes (Davis et al 2013, Lodree et al 2012, Lodree and Taskin 2009, Morrice et al 2016, Rawls and Turnquist 2010, Taskin and Lodree 2010, 2011). 1…”

Section: Introductionmentioning

confidence: 99%

Pre‐Hurricane Consumer Stockpiling and Post‐Hurricane Product Availability: Empirical Evidence from Natural Experiments

Pan¹,

Dresner

Mantin

et al. 2020

Production and Operations Management

View full text Add to dashboard Cite

T he provision of essential supplies is a key service provided by retailers when demand spikes due to consumer stockpiling during environmental emergencies. Moreover, it is important for retailers to quickly recover from these events by replenishing the stock of essential supplies to meet the continuing needs of local residents. The main purpose of this research is to study consumer precautionary stockpiling behavior prior to the onset of hurricane landfalls and determine the impact of this behavior on in-store product availability for various formats of retail store outlets. Specifically, we focus on the bottled water product category, an essential emergency category in hurricane preparedness. This study combines an event analysis methodology with econometric models using archival retail scanner data from 60 US retail chains located in 963 counties and real-time data from four recent US continental hurricanes. We find that supply-side characteristics (retail network and product variety), demand-side characteristics (hurricane experience and household income), and disaster characteristics (hazard proximity and hazard intensity) significantly affect consumer stockpiling propensity as the hurricanes approach. The increased consumer stockpiling has immediate and longer-term impacts on retail operations, namely, in-store product availability. Among various retail formats, drug stores are associated with the highest consumer stockpiling propensity before hurricanes, while dollar stores and discount stores are associated with the lowest in-store product availability following hurricanes. Our study points to the need for retailers and policymakers to carefully monitor factors affecting consumer stockpiling behavior that will allow for better allocation of critical supplies during the hurricane season.

show abstract

Section: Introductionmentioning

confidence: 99%

Pre‐Hurricane Consumer Stockpiling and Post‐Hurricane Product Availability: Empirical Evidence from Natural Experiments

Pan¹,

Dresner

Mantin

et al. 2020

Production and Operations Management

View full text Add to dashboard Cite

show abstract

“…The authors approach the problem as an optimal stopping problem with Bayesian updates. The paper given by Taskin and Lodree (2011) extends their prior work by considering a multi-retailer supply chain and by incorporating official forecasts from the NHC.…”

Section: Literature Reviewmentioning

confidence: 91%

“…A major drawback from the papers given by Lodree and Taskin (2009), and Taskin and Lodree (2011) is that they assume statistical independence of sequential forecast advisories. This assumption seems unrealistic and would need to be verified in order to guarantee the applicability of their study under real-life settings.…”

Section: Literature Reviewmentioning

confidence: 99%

“…In their paper, the authors develop a model for establishing the optimal evacuation time for a given household when a hurricane is approaching. As in Lodree and Taskin (2009), and Taskin and Lodree (2011), the objective is to determine when to stop taking observations, and then start the evacuation process. In Czajkowski and Woodward (2010), the authors model the hurricane forecasts as a Markovian process that assumes that the forecast in the next time period depends only on the current forecast and not on the previous.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Forecast-driven model for prepositioning supplies in preparation for a foreseen hurricane

Pacheco

Batta

2016

Journal of the Operational Research Society

View full text Add to dashboard Cite

In this paper, we present a forecast-driven dynamic model for prepositioning relief items in preparation for a foreseen hurricane. Our model uses forecast advisories issued by the National Hurricane Center (NHC), which are issued every 6 h. Every time a new advisory is issued with updated information, our model determines the amount and location of units to be prepositioned and it also re-prepositions already prepositioned units. The model also determines the best time for starting the prepositioning activities. Our approach uses a combination of Decision Theory and stochastic programming. The outcomes of our model are presented in a way that could be easily understood by humanitarian practitioners who are ultimately the ones who would use and apply our model.

show abstract

“…Other social aspects associated with the occurrence of Hurricane have also been reported. For instance, a Bayesian decision model for management of emergency supplies has been presented by Taskin and Lodree [155]. Guikema et al [156] presented a model for predicting power outages associated with hurricanes.…”

Section: Hurricanesmentioning

confidence: 99%

Current Efforts for Prediction and Assessment of Natural Disasters: Earthquakes, Tsunamis, Volcanic eruptions, Hurricanes, Tornados, and Floods

2017

View full text Add to dashboard Cite

Abstract. This article presents a state-of-the-art review of di erent methods, signal and image processing techniques, and statistical analyses used for prediction and assessment of natural disasters including earthquakes, tsunamis, volcanic eruptions, hurricanes, tornadoes, and oods. Application of the big data paradigm to the aforementioned natural disasters is also discussed. The research for increasingly more sophisticated computational models will continue to achieve more accurate predictions of various natural disasters.

show abstract

A Bayesian Decision Model with Hurricane Forecast Updates for Emergency Supplies Inventory Management

Cited by 35 publications

References 23 publications

Pre‐Hurricane Consumer Stockpiling and Post‐Hurricane Product Availability: Empirical Evidence from Natural Experiments

Pre‐Hurricane Consumer Stockpiling and Post‐Hurricane Product Availability: Empirical Evidence from Natural Experiments

Forecast-driven model for prepositioning supplies in preparation for a foreseen hurricane

Current Efforts for Prediction and Assessment of Natural Disasters: Earthquakes, Tsunamis, Volcanic eruptions, Hurricanes, Tornados, and Floods

Contact Info

Product

Resources

About