Multi-criteria decision-making for complex bundling configurations in surface transportation of air freight
Introduction
Air cargo, which is defined as any property (e.g.: freight, mail, express parcels) transported by a full-freight aircraft, a combi-aircraft, or under the main deck of a passenger aircraft (Domingues et al., 2014), has at least two advantages over surface transportation (sea, road and rail), the first of which is the speed of the transportation, which is required for specific goods, such as perishable goods or goods that require next morning delivery (e.g. newspapers). The second advantage is the low risk for damage or loss, which means that high-value, time-sensitive goods with a high value-to-weight ratio are suitable for air transportation (Zhang and Zhang, 2002, Ohashi et al., 2005).
The air freight transport chain has different stages, starting at the shipper, leading via forwarders and the carrier (in our case the airline) towards the consignee. The logistics chain of air freight is visualized in Fig. 1 (adopted from Petersen, 2007). Shippers, the clients of the air cargo transporting service, are positioned at the start of the logistics chain (Popescu et al., 2010). They use freight forwarders to transport their shipment to a carrier. Other services being offered by freight forwarders include securing freight, storing freight, consolidating freight, organizing value added services and loading freight trucks. The carrier moves the shipment with airplanes to the respective destination airport, from which forwarders transport and deliver the freight shipment to the consignee, the recipient of the shipment (Petersen, 2007).
Fig. 2 shows an example of a high-level overview of the air freight transport chain. Usually, carriers contract Third Party Logistics Service providers (3PL) in the country or continent of origin to collect freight from freight forwarders and send it to the airport, together with other shipments. The same typically applies to the destination airport, but then in reverse order, as outlined in Fig. 2.
Here we briefly describe the actors involved in this chain:
Shipper: Shipper is the owner or supplier of the freight. It is also called ‘consignor’, its ultimate goal is to send shipment to the ‘consignee’.
Forwarder: Forwarder is the actor that arranges the transport of the freight from shipper to origin outstation or from destination airport to consignee. Forwarders often consolidate shipments, so they combine shipments designated for the same flight, in order to get a better price from the carrier. They also provide additional services, such as custom declarations and registrations of dangerous goods.
Outstation: Outstation is a place where an airline receives freight from forwarders or shippers. Outstation is usually operated by Third Party Logistics Service providers (3PL). Airlines usually contract these 3PL companies to process freight for airports.
Road carrier: A road carrier is a trucking company which transports freight from outstation to hub under an airway bill. Air carriers usually outsource their trucking activities to trucking companies. These trucks function as an airline service between outstations and the hub. These trucks comply with all requirements to transport air freight.
Hubs: Hubs receive freight from either trucks originated from outstations or directly from the shipper. Once arrived and accepted in the hub, the palletized freight is stored, ULD's are decomposed to shipment level or (re)built on pallets for the correct. The main goal of the hub is that the shipments are built on the correct ULDs and are ready for carriage before the deadline of the flight.
Air carrier: Air carrier is the airline on which the ULDs are loaded and transported to the destination.
Consignee: Consignee is the final receiver of the shipment.
Air freight transport chain might be of some other forms. What we presented above, is the framework we use in our study.
To transport air freight to an airport hub it needs to be palletized on unit load devices (ULDs). The most common ULDs are containers and big metal plates, on which freight can be bundled and tied down. For transportation by truck, the 3PL typically has three options in terms of the transport of shipments to an airport:
- -
Through-ULD (T-ULD); this type of ULD contains shipments for the same flight from an air cargo hub. This type is desirable for carriers, but only possible when there is enough freight volume at the outstation for a specific destination to build T-ULDs.
- -
Mixed-ULD (M-ULD); ULD that contains shipments for multiple flights.
- -
Loose freight; instead of palletizing freight, shipments are placed on a skid inside a truck.
T-ULDs are restricted by the specifications of the aircraft: if, for example, only belly freight can be accommodated on an aircraft, which is the primary form of carrying air cargo at many large airports (Merkert and Ploix, 2014), the T-ULD pallet cannot exceed 10 m3 in volume. M-ULDs, on the other hand, are not restricted by aircraft specifications, because these pallets are disassembled upon arrival at the airport hub, as they do not contain shipments for one and the same flight. These ULDs can contain approximately 15 m3 of freight during transport by truck towards the hub. Transporting palletized shipments has two advantages: (i) it reduces the time needed to load and unload the trucks, and (ii) it lowers transport and storage cost, because palletized freight uses space more effectively. Disadvantages of transporting palletized shipments are: (i) the costs of palletizing freight, (ii) the need for moving equipment between hub and outstation, and (iii) the requirements of this equipment (Morabito et al., 2000). M-ULDs have an additional disadvantage, in that they also need to be disassembled upon arrival at the airport, because they contain shipments for multiple flights.
Carriers have three different optimization alternatives at their disposal when it comes to transporting palletized freight from a 3PL to an airport. These optimization alternatives apply to a situation in which all freight is palletized. These alternatives are:
- 1)
Increase the number of T-ULDs for non-constraint destinations: lower the minimum required volume for building T-ULDs at the outstation for shipments to non-constraint destinations. Because the freight capacity of flights to non-constraint destinations is not completely utilized, it is not necessary to build T-ULDs that utilize all the available cargo space of the ULD. This configuration will increase the number of T-ULDs and reduce the number of M-ULDs.
- 2)
Transport shipments loose: shipments with short connection time that are placed on a M-ULD can also be transported loose in trucks instead of palletized. Upon arrival at the airport hub, loose shipments can immediately be placed on a build-up buffer. This configuration increases the number of loose shipments and reduces the number of M-ULDs.
- 3)
Optimize trucks: This configuration is aimed at utilizing all ULD capacity to minimize trucking cost. Therefore, for instance, instead of building lower deck T-ULDs of 10 m3, M-ULDs of 15 m3 are built at outstations.
The aim of this study is to identify the optimal strategy for planning freight for transport from outstation to a hub. A multi-criteria decision-making problem is formulated and solved to determine the optimal freight bundling configuration for transport. It is, therefore, important to identify the key factors for selecting the right freight bundling configuration. This study only focuses on the factors that either relate to freight bundling configurations or could impact the optimal configuration indirectly. The scope of this research comprises all the processes from the moment freight is delivered by the freight forwarder to the 3PL to the point that shipments are placed on T-ULDs before flights at an airport freight hub.
This study offers the following contributions. The first contribution is the structured approach designed to model a complex freight system. This model shows the impact of changing the freight bundling configuration on the basis of the KPIs of a logistics service provider in the air cargo sector. The model was designed, verified, validated and tested at KLM Cargo. One of the strengths of this study is that the scope not only includes multiple actors from inside a freight carrier, but also external actors, like truck operators and outstations. This provides a more complete overview of the impact of certain freight bundling configurations. A second contribution of this research is the identification of the key factors in determining the best strategy for outstations that transport shipments to air freight hubs. For this purpose, we use a new multi-criteria decision-making method called best worst method (BWM). While most existing studies focus on one aspect of the problems of this kind, this study takes multiple KPIs into account, which avoids sub-optimality.
The next section contains a literature review involving existing studies on this topic and on the relevant KPIs used in the transport and logistics industry. The methodology is discussed in section 3, followed by an application of this methodology on case study on KLM Cargo in section 4. Section 5 discusses the conclusion and suggestions for future research.
Section snippets
Literature review
There is a limited number of studies available regarding the optimization of air freight, a subject that is discussed in section 2.1. A literature review on the important KPIs in the transportation industry is presented in section 2.2.
Mathematical modeling
In this section, we first present the notations and assumptions used in the remainder of this section, and then discuss the mathematical modeling of the problem.
Application
In this section, the methodology discussed above is applied in a case study involving KLM Cargo.
Conclusions and future research
This study presents a methodology to test various alternatives for optimizing an export air freight chain, which includes three bundling freight options: on T-ULD (all freight for the same flight), on M-ULD (freight for different flights), and loose freight in trucks. In such a situation, the goal for a carrier is to find the best freight bundling configuration alternative. We first formulate the problem as a multi-criteria decision-making (MCDM) problem considering three KPIs (criteria):
References (23)
- et al.
Optimising transport efficiency and costs in Australian wood chipping operations
Biomass Bioenergy
(2012) - et al.
The potential of horizontal collaboration in airport ground freight services
J. Air Transp. Manag.
(2014) - et al.
An assessment of the regulation of air cargo security in Europe: a Belgian case study
J. Air Transp. Manag.
(2014) - et al.
An AI approach for optimizing multi-pallet loading operations
Expert Syst. Appl.
(2009) - et al.
The impact of terminal re-organisation on belly-hold freight operation chains at airports
J. Air Transp. Manag.
(2014) - et al.
Loading optimization of palletized products on trucks
Transp. Res. Part E Logist. Transp. Rev.
(2000) - et al.
Choice of air cargo transshipment airport: an application to air cargo traffic to/from Northeast Asia
J. Air Transp. Manag.
(2005) Best-worst multi-criteria decision-making method
Omega
(2015)- et al.
An efficient approach for the multi-pallet loading problem
Eur. J. Oper. Res.
(2000) - et al.
Real time simulation of auctioning and re-scheduling processes in hybrid freight markets
Transp. Res. Part B Methodol.
(2007)
Network planning for scheduling operations in air cargo handling: a tool in medium term goods flow control
Eur. J. Oper. Res.
Cited by (106)
Performance optimization of fourth-generation biorefinery departments: A novel mixed macroergonomics-sustainability framework
2024, Sustainable Materials and TechnologiesAn integrated hybrid MCDM approach to evaluate countries’ COVID-19 risks
2023, Socio-Economic Planning SciencesA neighborhood-level analysis of association between social vulnerability and COVID-19 in ahvaz, Iran
2023, International Journal of Disaster Risk ReductionDifferences in perception of the importance of process safety indicators between experts in Iran and the West
2023, Journal of Safety ResearchA new risk assessment framework for safety in oil and gas industry: Application of FMEA and BWM based picture fuzzy MABAC
2022, Journal of Petroleum Science and EngineeringPromoting Australian regional airports with subsidy schemes: Optimised downstream logistics using vehicle routing problem
2022, Transport PolicyCitation Excerpt :The results indicate that Changi Airport is the most competitive airport in terms of airport facilities and operations, while Hong Kong has the number one ranked airport in terms of airfreight capacity and economic development (see also Zhang (2003), for an in-depth analysis of Hong Kong’s competitiveness in air cargo). Rezaei et al. (2017) used a multi-criteria tool to determine the optimal strategy for freight shipment from outstations to hub airports with a consideration of the cost, loading time and quality. They found that vehicles costs and freight handling charges are the main factors in determining the optimal freight bundling configuration.