Robust spotter scheduling in trailer yards
- PDF / 2,573,484 Bytes
- 27 Pages / 439.37 x 666.142 pts Page_size
- 101 Downloads / 123 Views
Robust spotter scheduling in trailer yards Giorgi Tadumadze1 · Nils Boysen2 · Simon Emde3 Received: 4 August 2019 / Accepted: 23 July 2020 © The Author(s) 2020
Abstract Spotters (also denoted as switchers) are specialized terminal tractors, which are dedicated to the rapid maneuvering of semitrailers between parking lot and dock doors in large trailer yards. This paper is dedicated to spotter scheduling, i.e., the assignment of predefined trailer movements to a given fleet of spotters. The limited number of dock doors for loading and unloading is often the scarce resource dur‑ ing trailer processing, so that idle time of the bottleneck, e.g., caused by unfore‑ seen delay in the yard, is to be avoided. In this setting, we aim to insert time buff‑ ers between any pair of subsequent jobs assigned to the same spotter, so that small delays are not propagated and subsequent jobs can still be executed in a timely man‑ ner. We formalize two versions of the resulting robust spotter scheduling problem and provide efficient algorithms for finding optimal solutions in polynomial time. Furthermore, we simulate delays during the execution of spotter schedules and show that the right robustness objective can greatly improve yard performance. Keywords Yard operations · Truck scheduling · Terminal tractor scheduling · Robustness
* Nils Boysen nils.boysen@uni‑jena.de http://www.om.uni-jena.de Giorgi Tadumadze giorgi.tadumadze@tu‑darmstadt.de http://www.or.wi.tu-darmstadt.de Simon Emde [email protected] https://pure.au.dk/portal/en/[email protected] 1
Fachgebiet Management Science/Operations Research, Technische Universität Darmstadt, Hochschulstraße 1, 64289 Darmstadt, Germany
2
Lehrstuhl für Operations Management, Friedrich-Schiller-Universität Jena, Carl‑Zeiß‑Straße 3, 07743 Jena, Germany
3
Department of Economics and Business Economics, Aarhus University, Fuglesangs Allé 4, 8210 Aarhus V, Denmark
13
Vol.:(0123456789)
G. Tadumadze et al.
1 Introduction Increasing freight traffic in many regions of the world (e.g., in Europe Statista 2018) does not only burden the edges (streets) of road networks but also many nodes. Examples of huge terminals where up to several hundred trucks are to be loaded and unloaded each day are cross docks (Ladier and Alpan 2016), automobile plants (Bat‑ tini et al. 2013), distribution centers of food retailers (Bodnar et al. 2015), freight airports (Ou et al. 2010), and hub terminals in the postal service industry (Boysen et al. 2017). In many of these nodes, especially during peak hours, considerable waiting times for an (un-)loading of trucks occur. The large German transport coop‑ erative Elvis with 10,643 trucks, for instance, reports that their average daily waiting times of 3.5 h per truck accumulate to yearly waiting costs of about €400 million (VRS 2012). In an empirical study, 45.9% of the surveyed German truck drivers quantify their average waiting time per stop to exceed 1 h (VRS 2011). Existing ideas on how to resolve this problem mainly address the demand side. Novel soft‑
Data Loading...
