TY - JOUR
T1 - Effects of application costs on fertilizer application strategy
AU - Tröster, Michael Friedrich
AU - Pahl, Hubert
AU - Sauer, Johannes
N1 - Publisher Copyright:
© 2019 Elsevier B.V.
PY - 2019/12
Y1 - 2019/12
N2 - To optimize production activities, it is important to understand the associated costs. If the optimization is carried out using mathematical instruments, the production costs are implemented in the form of a restriction. The functional form is critical not only to ensure accuracy but also to facilitate computing power and input data requirements. The present study documents the development of a cost function for fertilizer application. Three potential ways to address transportation costs within the whole cost function are observed: (i) calculating minimal transportation time using a “split delivery vehicle routing problem” (SDVRP), (ii) estimating transportation time using a regression model, and (iii) neglecting transport costs altogether. In Section 3, the costs of fertilizer application and their influence on the fertilizer application strategy are compared. Despite minimal differences in the application cost values, all methods lead to comparable results. A further investigation reveals additional factors that influence the reliability of decision-making, for example, price relations. The computational power and data input demands are explored as well. In this respect, the SDVRP method was identified as the most resource-demanding option. We conclude that the performance of the regression method is the most reliable for optimizing the fertilizer application strategy using mathematical instruments. The present study may support researchers focusing on farm logistics or related cost functions, such as spraying, sowing or manure application.
AB - To optimize production activities, it is important to understand the associated costs. If the optimization is carried out using mathematical instruments, the production costs are implemented in the form of a restriction. The functional form is critical not only to ensure accuracy but also to facilitate computing power and input data requirements. The present study documents the development of a cost function for fertilizer application. Three potential ways to address transportation costs within the whole cost function are observed: (i) calculating minimal transportation time using a “split delivery vehicle routing problem” (SDVRP), (ii) estimating transportation time using a regression model, and (iii) neglecting transport costs altogether. In Section 3, the costs of fertilizer application and their influence on the fertilizer application strategy are compared. Despite minimal differences in the application cost values, all methods lead to comparable results. A further investigation reveals additional factors that influence the reliability of decision-making, for example, price relations. The computational power and data input demands are explored as well. In this respect, the SDVRP method was identified as the most resource-demanding option. We conclude that the performance of the regression method is the most reliable for optimizing the fertilizer application strategy using mathematical instruments. The present study may support researchers focusing on farm logistics or related cost functions, such as spraying, sowing or manure application.
KW - Application costs
KW - Fertilizer strategy
KW - Route-planning
KW - SDVRP
KW - Transport costs
UR - http://www.scopus.com/inward/record.url?scp=85072960543&partnerID=8YFLogxK
U2 - 10.1016/j.compag.2019.105033
DO - 10.1016/j.compag.2019.105033
M3 - Article
AN - SCOPUS:85072960543
SN - 0168-1699
VL - 167
JO - Computers and Electronics in Agriculture
JF - Computers and Electronics in Agriculture
M1 - 105033
ER -