Value of Image-based Yield Prediction: Multi-location Newsvendor Analysis

被引:1
作者
Amaruchkul, Kannapha [1 ]
机构
[1] Natl Inst Dev Adm NIDA, Grad Sch Appl Stat, Logist Management Program, 118 Serithai Rd, Bangkok 10240, Thailand
来源
OPERATIONS RESEARCH AND ENTERPRISE SYSTEMS, ICORES 2019 | 2020年 / 1162卷
关键词
Image processing; Agricultural supply chain; Applied operations research; Stochastic model applications; MULTIPRODUCT NEWSVENDOR; MODELS; INFORMATION; INTERNET;
D O I
10.1007/978-3-030-37584-3_1
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
Consider an agricultural processing company, which wants to pre-purchase crop from different locations before a harvesting season in order to maximize the total expected profit from all outputs subject to multiple resource constraints. The yields for different outputs are random and depend on the location. By remotely sensing from satellites or locally sensing from unmanned aerial vehicle, the firm may employ an image-based yield prediction model at the pre-purchase time. The distribution of the yield differs by a location. With the sensed data, the company updates the distribution of the yield using a regression model, whose explanatory variable is a vegetation index from image processing. At a more favorable location, the distribution of the yield is stochastically larger. The objective of this paper is to quantify the added value of image sensing in predicting crop yield. Specifically, the posterior yield distribution from image processing is used as an input to the multi-location newsvendor model with random yields. The optimal expected profit given the posterior distribution is compared to that with only the prior distribution of the yield. The difference between the total expected profits with the prior and posterior distributions is defined as the value of the sample information. We derive the type-1 and -2 errors as a function of the standard error of the estimate. In the numerical example, we show that the value of the sample information tends to be increasing (with diminishing return) as the yield prediction model becomes more accurate.
引用
收藏
页码:3 / 22
页数:20
相关论文
共 34 条
[1]   A quadratic programming approach to the multi-product newsvendor problem with side constraints [J].
Abdel-Malek, Layek L. ;
Areeratchakul, Nathapol .
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 2007, 176 (03) :1607-1619
[2]   Application of planning models in the agri-food supply chain: A review [J].
Ahumada, Omar ;
Villalobos, J. Rene .
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 2009, 196 (01) :1-20
[3]  
Amaruchkul K, 2019, P 8 INT C OPERATIONS
[4]  
[Anonymous], 2012, Probability and Statistics
[5]  
[Anonymous], 2002, Wiley Series in Probability and Statistics
[6]   Integrating remote sensing, census and weather data for an assessment of rice yield, water consumption and water productivity in the Indo-Gangetic river basin [J].
Cai, X. L. ;
Sharma, B. R. .
AGRICULTURAL WATER MANAGEMENT, 2010, 97 (02) :309-316
[7]   Water productivity and crop yield: A simplified remote sensing driven operational approach [J].
Campos, Isidro ;
Neale, Christopher M. U. ;
Arkebauer, Timothy J. ;
Suyker, Andrew E. ;
Goncalves, Ivo Z. .
AGRICULTURAL AND FOREST METEOROLOGY, 2018, 249 :501-511
[8]   On the multi-product newsvendor with bounded demand distributions [J].
Chernonog, Tatyana ;
Goldberg, Noam .
INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS, 2018, 203 :38-47
[9]  
Choi S., 2012, HDB NEWSVENDOR PROBL, V176, DOI [10.1007/978-1-4614-3600-3_2, DOI 10.1007/978-1-4614-3600-3_2]
[10]  
Choi T M., 2012, Handbook of Newsvendor Problems, International Series in Operations Research and Management Science, International Series in Operations Research Management Science, DOI DOI 10.1007/978-1-4614-3600-3