Internet-of-Things-augmented dynamic route planning approach to the airport baggage handling system

https://doi.org/10.1016/j.cie.2022.108802Get rights and content

Highlights

  • Propose a dynamic route planning approach to the airport baggage handling systems.

  • Exploit real-time information acquired with IoT technologies.

  • Verify the proposed solution with an actual airport case.

  • Compare the results with traditional decentralized heuristic algorithm.

  • Can help maintenance by identifying susceptible transmission links.

Abstract

This paper proposes a dynamic route planning approach to the airport baggage handling systems. In previous studies, when planning the baggage transmission routes, real-time conditions of the baggage handling system are often neglected. This study incorporates the real-time baggage tracking information and device disruption information into the baggage route planning algorithm based on Internet-of-Things technologies. Taking Hierarchical Cooperative A* algorithm (HCA*) as the basic routing core, we propose a comprehensive algorithmic framework for the dynamic baggage route planning problem, in order to iteratively generate conflict-free baggage transmission routes while adapting to varying system conditions. We propose a unified prioritizing scheme so that baggage can be handled in proper order according to their arrival, departure and waiting time. This solution approach also contains a disruption handling mechanism by which the affected transportation links and baggage can be appropriately handled. A real airport case is used to verify the proposed solution approach. The results show that our proposed method is around 10% more efficient than the decentralized heuristic algorithm. Meanwhile, the average baggage throughput time decreases about 5% when real-time information is included. This paper also shows how to identify susceptible transmission links so specific maintenance plans can be arranged.

Introduction

With the continuing growth in worldwide passenger numbers and ongoing globalization, the number of baggage handled daily at airports is steadily increasing (Abdelghany et al., 2006). The baggage that needs to be loaded onto the aircraft has to undergo more than ten processes, among them, scanning, screening, transportation, sorting, transit, loading, unloading, and delivery, involving multiple technical fields (Johnstone et al., 2010). A baggage handling system (BHS) is a system that supports such processes, transportation of checked baggage from ticket counters to areas where the bags can be loaded onto aircrafts. A BHS also transports checked baggage coming from airplanes to baggage claims or to an area where the bag can be loaded onto another airplane. The total length of the baggage transportation lines in a medium-size airport is typically up to tens of kilometers, making it one of the largest and most complex single systems in the terminal (De Neufville, 1994). The steadiness, reliability and efficiency of the BHS are crucial to the operation of the terminal.

BHS applications can be traced back to the late 1960s (Robinson, 1969). In general, there are three BHS types: conveyor-based system, merging-conveyor system, and DCV (Destination Coded Vehicle) system. A conveyor-based system is the conventional one that guides the baggage through a fixed path. The merging-conveyor system, also called Individual Carrier System (ICS), adopts a new automatic baggage handling mode of “one baggage and one tote” to achieve high baggage handling efficiency. ICS adopts the “simple carrier–complex track” approach (Sørensen et al., 2020) by which a “simple” tote can carry a bag, and run on a “complex” belt conveyor system. Compared to conventional conveyor-based systems, the ICS system has controllable diverting and merging functions and redundant paths in the conveyor network, which can adapt to changes in traffic flow, therefore facilitating more flexibility and reliability in material handling. The DCV systems adopt a “complex carrier–simple track” pattern that carriers can maneuver by themselves along with the rack. Cutting-edge DCV systems use Automatic Mobile Robots (AMR) as carriers that can move without relying on tracks. Since ICS and DCV are the mainstream BHS technologies, and both facilitate the appropriate routing of baggage, we focus on the routing problem of ICS and DCV in this study.

For a BHS system, the routing problem is a significant control challenge (Zeinaly et al., 2015). It involves routing loaded totes from loading stations to unloading stations and routing empty totes from unloading stations to loading stations. A loading station is where bags are loaded onto totes after entering the system (either from check-in counters or from transfer flights). The unloading station is where the bags are unloaded before loading them onto the aircraft. If a bag enters the system too early, the BHS will guide it to an early baggage storage (EBS) entrance. The EBS is typically an automated storage/retrieval system used to temporarily store the loaded totes. Loading stations, unloading stations, and EBSs are connected by a network of unidirectional tracks on which the totes travel.

The traditional approaches to control a BHS have focused on directing baggage along a prior computed (static) shortest route (Hallenborg & Risager, 2007). Some recent studies have proposed dynamic routing approaches which can handle gradually revealed task information (Tarău et al., 2010, Zeinaly et al., 2015). However, those approaches, for example, the nonlinear non-convex programming solvers or mixed-integer linear programming (MILP) solvers, are very time-consuming; therefore, they were only tested on toy instances with no more than 10 junctions, far away from practical application. Another research gap lies in the traffic condition of the system, which is not explicitly considered in previous studies. In practice, BHS is subject to unpredictable events and mechanical control errors which may cause congestion or conflict, thus decreasing the capability of the BHS. For example, a failure of one short link may lead to a wide-range disruption since more and more bags are approaching this link if no adjustment is taken, leading to congestion on the link before the disrupted one. Another potential conflict comes from the error between the scheduled baggage trajectory and the real one since the baggage speed can differ with what we have planned due to the mechanical control precision. Thus, if the real traffic on the links deviates from the planned situation, the current baggage routes are subject to an adjustment in pursuit of better performance. In the era of the Internet of Things (IoT), it is possible to sample real-time baggage tracking information (BTI) and device disruption information (DDI). BTI data can be represented by (e,l,t) where e is a link, l is the length from the origin of link l to the bag’s position, and t is the current time. BTI data can be captured by the sensors deployed along with the links. DDI can be represented by (e,s) where s indicates whether link e is disrupted. DDI can be detected by some monitoring sensors. To the best of our knowledge, none of the prior studies have incorporated such information in the route planning of airport baggage.

In this study, we propose an IoT-augmented dynamic route planning approach (IoT-DRPA) for BHS. This approach can create routing plans for new-arrival baggage while re-planning routes for baggage with potential conflicts with others. The contribution of this study lies in three aspects.

(1) We proposed a new time-affordable centralized route planning method based on the HCA* algorithm. While in contrast, prior studies use decentralized heuristics or centralized LP-based approximation methods to reduce CPU time. Computational results validate that our method outperforms the decentralized one (adapted from Tarău et al., 2009b) by around 10% regarding the average throughput time.

(2) We incorporated the real-time information acquired with IoT technologies to hedge against the unpredictable device disruption and control error and proposed a dynamic route planning framework.

(3) We studied a real case of an airport baggage handling system with a scale that has not been studied before. The case contains 7 loading stations, 22 unloading stations, and 44 junctions capable of handling over 28,000 bags per day. The computational results validate that the proposed method meets the requirement of baggage handling in medium-scale airports.

The rest of this paper is organized as follows. In Section 2, we present a review of the state-of-art researches in the field, followed by the definition of the problem in Section 3. In Section 4, we elaborate on the proposed dynamic route planning approach. In Section 5, we set up experiments and analyze the results. Section 6 concludes this paper and points out possible directions for future research.

Section snippets

Literature review

This section first reviews direct solution approaches to BHS route planning; then we summarize the algorithms of multi-agent pathfinding (MAPF) problems which we believe are relevant.

Problem description

The main process of the outbound baggage handling, as shown in Fig. 1, starts from the check-in counter and ends at the sorting area where bags are sorted into their corresponding aircrafts. The inbound baggage handling process is likewise and is ignored here.The outbound side of the baggage handling system is to sort the claim baggage towards the correct baggage claim carousel or sort the transfer baggage towards designated outbound stations (Boysen et al., 2019). The first case is trivial

The dynamic route planning approach

In this paper, we present an IoT-augmented dynamic route planning approach (IoT-DRPA) to deal with the BHS routing problem. We first introduce the overall framework of the proposed approach, then we elaborate the resolution of device disruptions and the unified prioritizing scheme.

Case study

In this section, we implement the IoT-DRPA approach with a real case of an airport. There are two main purposes for this case study: to evaluate the performance of the proposed solution approach and to shed light on administrative insights. A simulation system is developed from scratch to implement the process illustrated in Fig. 3. The system runs on a computer with Intel(R) Core(TM) i7-8550U CPU at 1.8G and 2.0G and 8 GB memory.

Conclusion

In this paper, we propose an IoT-augmented dynamic route planning approach, called IoT-DRPA, to solve baggage routing problems in airports. Two kinds of IoT information are adopted to improve the efficiency of the BHS: the baggage tracking information (BTI) and the device disruption information (DDI). Such information enables the route planning system to identify and forecast traffic conflicts in the conveying network and allows it to re-route affected baggage in real time. It also helps

CRediT authorship contribution statement

Xiuqing Yang: Conceptualization, Methodology, Funding acquisition, Project administration. Ruchen Feng: Software, Visualization, Writing – original draft. Pengcheng Xu: Investigation, Data curation. Xiaorui Wang: Investigation, Data curation. Mingyao Qi: Conceptualization, Methodology, Validation, Writing – review & editing, Funding acquisition, Supervision.

Acknowledgments

This work is supported by Sichuan Science and Technology Program under grant No. 21ZDYF3628, 2021JDRC0009 and the National Natural Science Foundation of China under grant No. 71772100.

References (30)

  • GradyD.K. et al.

    Asynchronous distributed motion planning with safety guarantees under second-order dynamics

  • HallenborgK. et al.

    Decide: agents controlling a BHS of an airport hub

    Journal of Automation Mobile Robotics and Intelligent Systems

    (2007)
  • HartP.E. et al.

    A formal basis for the heuristic determination of minimum cost paths

    IEEE Transactions on Systems Science and Cybernetics

    (1968)
  • HolkarK. et al.

    An overview of model predictive control

    International Journal of Control and Automation

    (2010)
  • JohnstoneM. et al.

    Status-based routing in Baggage handling systems: Searching verses learning

    IEEE Transactions on Systems, Man and Cybernetics, Part C. Applications and Reviews: A Publication of the IEEE Systems, Man, and Cybernetics Society

    (2010)
  • Cited by (0)

    This work is supported by Sichuan Science and Technology Program under grant No. 21ZDYF3628, 2021JDRC0009 and the National Natural Science Foundation of China under grant No. 71772100.

    View full text