Abstract Statistics about the efficiency of multi-machinery operations can provide valuable support to machinery managers for decision making, and benchmarks for the evaluation of computer-based optimizations of the operations. However the acquisition of data from the operations execution requires a high level of manual invention. This paper presents algorithmic methods for automatic recognition of machine operation modes during grain harvest operations involving multiple combines and multiple transport units by analyzing recorded GNSS-trajectories. The developed methods infer the operation modes of the machines using simple models of how grain harvesting is conducted. The model parameters and prediction performance of the methods were evaluated on the basis of experimental data from a real grain harvest operation. As no ground truth was available the correct functioning of the methods had to be evaluated on the basis of expected results based on knowledge of the grain harvesting process. The evaluation did however indicate that the developed methods perform well.
[1]
Claus G. Sørensen,et al.
Operational Analyses and Model Comparison of Machinery Systems for Reduced Tillage
,
2005
.
[2]
Claus G. Sørensen,et al.
The vehicle routing problem in field logistics: Part II
,
2009
.
[3]
David H. Douglas,et al.
ALGORITHMS FOR THE REDUCTION OF THE NUMBER OF POINTS REQUIRED TO REPRESENT A DIGITIZED LINE OR ITS CARICATURE
,
1973
.
[4]
Henry A. Kautz,et al.
Learning and inferring transportation routines
,
2004,
Artif. Intell..
[5]
J. Munkres.
ALGORITHMS FOR THE ASSIGNMENT AND TRANSIORTATION tROBLEMS*
,
1957
.
[6]
Michael Beetz,et al.
Computerized real-time analysis of football games
,
2005,
IEEE Pervasive Computing.
[7]
Robert D. Grisso,et al.
ANALYSIS OF TRAFFIC PATTERNS AND YIELD MONITOR DATA FOR FIELD EFFICIENCY DETERMINATION
,
2002
.