Real-time container transport planning with decision trees based on offline obtained optimal solutions

Abstract

Hinterland networks for container transportation require planning methods in order to increase efficiency and reliability of the inland road, rail and waterway connections. In this paper we aim to derive real-time decision rules for suitable allocations of containers to inland services by analysing the solution structure of a centralised optimisation method used offline on historic data. The decision tree can be used in a decision support system (DSS) for instantaneously allocating incoming orders to suitable services, without the need for continuous planning updates. Such a DSS is beneficial, as it is easy to implement in the current practice of container transportation. Earlier proposed centralised methods can find the optimal solution for the intermodal inland transportation problem in retrospect, but are not suitable when information becomes gradually available. The main contributions are threefold: firstly, a structured method for creating decision trees from optimal solutions is proposed. Secondly, an innovative method is used for obtaining multiple equivalent optimal solutions to prevent overfitting of the decision tree. And finally, a structured analysis of three error types is presented for assessing the quality of an obtained tree. A case study illustrates the method's purpose by comparing the quality of the resulting plan with alternative methods.

Dependability of the information flow process at an intermodal transhipment point

“…Informacja jest elementem, który wyprzedza przepływ ładunku, jednocześnie sygnalizując co, jak, gdzie i kiedy powinno się stać z danym ładunkiem. W przypadku obsługi ładunku w pracy [12] zostało poruszone zagadnienie planowania i optymalizacji obsługi kontenerów za pomocą metody drzew decyzyjnych. Modelowanie niezawodności w przypadku procesu transportowania w transporcie intermodalnym zostało przedstawione przykładowo w pracach [7,8,23].…”

“…Gdy kontener jest niezdatny, kontener jest odkładany na pole składowe i wykonywane są procedury wyjaśniające, co dalej zrobić z tym kontenerem, proces w tej chwili się kończy e 2 . Gdy kontener jest zdatny i nadana jest mu odpowiednia klasa, następuje przyjęcie ładunku na terminal poprzez wprowadzenie ładunku do systemu komputerowego d 12 . Następnie operatorzy maszyn odkładają ładunki d 13 na pole składowe, jeżeli ładunek nie został odłożony na plac składowy, to nie jest uznawane jako błąd.…”

“…If the container is up state, an appropriate class is assigned to it, and the load is accepted at the terminal by entering the load into the computer system d 12 . Next, machine operators place loads d 13 at the storage yard, if a load is not placed at the storage yard, it is not regarded as an error.…”

“…Information is an element that comes before the flow of load and signals, at the same time, what, how, where and when should happen to a given load. For load handling in study [12], the issue of planning and optimization of container handling using the decision tree method. Modelling of dependability in the case of the transporting process in intermodal transport is presented, for example, in studies [7,8,23].…”

Column generation based heuristic for learning classification trees

“…Decision trees are one of the most popular models for solving this problem, due to their effectiveness and high interpretability. They have been applied in a wide range of applications, from transport planning [34], internet advertisements [22], to healthcare interventions [24]. In this work, we focus on constructing univariate binary decision trees of prespecified depth.…”