Rollout-based routing strategies with embedded prediction: A fish trawling application

Fahrettin Cakir; Barrett W. Thomas; W. Nick Street

doi:10.1016/j.cor.2022.106055

Back

Rollout-based routing strategies with embedded prediction: A fish trawling application

Journal article

Peer reviewed

Rollout-based routing strategies with embedded prediction: A fish trawling application

Fahrettin Cakir, Barrett W. Thomas and W. Nick Street

Computers & operations research, Vol.150, p.106055

02/2023

DOI: 10.1016/j.cor.2022.106055

View Online

Abstract

In an effort to replenish stocks after years of overfishing, the European Union has been implementing policies to restrict the catch in all fisheries including in the Baltic Sea. These regulations incentivize fishermen to seek operational efficiency to counter lower catch limits. In this study, we address the fish trawling problem as a stochastic dynamic orienteering problem and propose rollout-based routing policies. Using a prescriptive analytic approach, we propose several prediction models and combine them with dynamic routing policies, the combination of which forms a prescriptive model. We evaluate and validate the effectiveness of these prescriptive models using a Baltic dataset on the fish species cod (Gadus). We propose a spatiotemporal cross-validation procedure to fairly assess different prescriptive models. Our findings show that reoptimization-based rollout strategies coupled with simple prediction methods such as nearest neighbors perform better than prescriptive models that use complex spatiotemporal smoothing techniques. •The Baltic cod trawling is modeled as a Markov Decision process.•Spatiotemporal correlation forecasting is used to generate rollout policies.•Spatiotemporal cross validation blocking is applied to assess model results.

Baltic cod fishery

Cross validation

Effort allocation

Fisheries management

Fishing behavior

Kriging

Prescriptive analytics

Rollout policies

Stochastic dynamic orienteering

Details

Title: Subtitle: Rollout-based routing strategies with embedded prediction: A fish trawling application
Creators: Fahrettin Cakir - İstanbul Sabahattin Zaim Üniversitesi
Barrett W. Thomas - Department of Business Analytics, University of Iowa, United States of America
W. Nick Street - Department of Business Analytics, University of Iowa, United States of America
Resource Type: Journal article
Publication Details: Computers & operations research, Vol.150, p.106055
Publisher: Elsevier Ltd
DOI: 10.1016/j.cor.2022.106055
ISSN: 0305-0548
eISSN: 1873-765X
Language: English
Date published: 02/2023
Academic Unit: Bus Admin College; Nursing; Computer Science; Business Analytics
Record Identifier: 9984380502702771

Metrics

6 Record Views