A stakeholder-based methodology to enhance the success of urban freight transport measures in a multi-level governance context

Research in Transportation Economics xxx (2017) 1e14

Contents lists available at ScienceDirect

Research in Transportation Economics journal homepage: www.elsevier.com/locate/retrec

A stakeholder-based methodology to enhance the success of urban freight transport measures in a multi-level governance context Bram Kin*, Sara Verlinde, Koen Mommens, Cathy Macharis MOBI e Mobility, Logistics and Automotive Technology Research Centre, Vrije Universiteit Brussel, Pleinlaan 2, Brussels, 1050, Belgium

a r t i c l e i n f o

a b s t r a c t

Article history: Received 13 October 2016 Received in revised form 18 January 2017 Accepted 7 February 2017 Available online xxx

Urban freight transport (UFT) is fundamental to the liveability of our cities, but it also contributes to the unsustainability of the same cities. Local authorities are primarily responsible for governing urban areas and implement different measures to regulate UFT. Measures often fail as they do not reach their intended goal and sometimes even produce adverse effects. One of the primary causes is that prior to implementation, the different stakeholders affecting and being affected by the UFT are not sufficiently involved. In this study, the multi-actor multi-criteria analysis (MAMCA) is applied ex-ante to evaluate to what extent different measures contribute to the objectives of different stakeholders. A municipality in Belgium, experiencing a lot of traffic, is used as a case study. Regulating UFT is in this case further complicated due to multi-level governance with different jurisdictions over infrastructure. To our knowledge, there are no similar UFT studies in a multi-level governance context. Results show that all proposed measures are an improvement for all stakeholders compared to the current situation. Future implementation of measures in this context is discussed. © 2017 Published by Elsevier Ltd.

JEL Classification: D70 R49 Keywords: Urban freight transport Sustainability Policy Stakeholder involvement MAMCA

1. Introduction The majority of the people, economic activities and consumption are concentrated in urban areas. A constant supply of goods is vital for the liveability of our cities (Lindholm, 2013). Consequently, a large number of freight vehicle movements is generated. Although freight vehicles only represent 8e15% of the total traffic flow, their share in transport-related emissions can be up to 50% (Dablanc, 2007). Freight vehicles contribute disproportionally to the unsustainability of cities with regard to economy (e.g., congestion), society (e.g., noise and unsafety) and the environment (e.g., air pollution) (MDS Transmodal, 2012; Quak, 2008). Nonetheless, only for the past years, the topic of urban freight transport (UFT) has been on the agenda of local authorities (Cherrett et al., 2012; Dablanc, 2007; Lindholm & Behrends, 2012; Lindholm, 2013; Stathopoulos, Valeri, & Marcucci, 2012). Local authorities regulate UFT with measures such as time windows, the provision of (un)loading zones, low emission zones,

* Corresponding author. E-mail addresses: [email protected] (B. Kin), [email protected] (S. Verlinde), [email protected] (K. Mommens), [email protected] (C. Macharis).

and weight and size restrictions (Anderson, Allen, & Browne, 2005; ~ uzuri, Larran ~ eta, Onieva, & Corte s, 2005). Based on the evaluMun ation of 106 possible measures, Quak (2008), nevertheless, concludes that they often do not reach their intended goal. One of the core reasons is that, the interests of different stakeholders in an urban (freight) context are not sufficiently taken into account in the decision-making process prior to implementation (Behrends, 2011; Bjerkan, Sund, & Nordtømme, 2014; Lindholm, 2013; Milan, Kin, ~ uzuri et al., 2005; Quak, 2008; Verlinde, & Macharis, 2015; Mun Stathopoulos et al., 2012; Witkowski & Kiba-Janiak, 2014). In this study, the multi-actor multi-criteria analysis (MAMCA) is applied to evaluate to what extent different UFT measures contribute to the objectives of stakeholders. The MAMCA explicitly incorporates the interests of different stakeholders in the decisionmaking process (Macharis, 2005). A case study is conducted in Mortsel, a municipality in the greater metropolitan area of Antwerp, in the Flanders region in Belgium. Mortsel experiences a lot of traffic. Decision-making for UFT is complicated due to multi-level governance with different jurisdictions (Marcucci & Stathopoulos, 2012; Marsden & Rye, 2010; Te Boveldt et al., 2016). Whereas the municipality is responsible for the sidewalks and parking, the regional government of Flanders is responsible for the provincial roads crossing through the city. The majority of the shops receiving

http://dx.doi.org/10.1016/j.retrec.2017.08.003 0739-8859/© 2017 Published by Elsevier Ltd.

Please cite this article in press as: Kin, B., et al., A stakeholder-based methodology to enhance the success of urban freight transport measures in a multi-level governance context, Research in Transportation Economics (2017), http://dx.doi.org/10.1016/j.retrec.2017.08.003

2

B. Kin et al. / Research in Transportation Economics xxx (2017) 1e14

freight flows is located along these roads. The added value of this case study is twofold. On the one hand, the authorities e local and regional e are supported in their decision-making process (aid decision-maker in understanding problem and possible alternatives) as well as in the decision outcome (aid in value judgements about trade-offs between conflicting objectives). The MAMCA as an extension of traditional multi-criteria decision analysis (MCDA) methods contributes to this by incorporating different interests, bringing stakeholders together, and identifying their advantages and disadvantages with regard to measures (Macharis, Turcksin, & Lebeau, 2012; Ward, Dimitriou, & Dean, 2016). A better understanding of the impact of measures on the different stakeholders enhances their success as potential bottlenecks are identified early in the process. On the other hand, the study contributes to UFT research. To our knowledge, there are no similar UFT studies in a multi-level governance context. Usually, solely the local authorities are taken into account (e.g., STRAIGHTSOL, 2014). However, as Lindholm (2012) mentions: ‘’The regional or state governments also affects the urban freight through overall policies and regulations that the local authorities need to consider, as well as there are, e.g. national road networks in some urban areas that are the responsibility of regional or state governments, …’’ (p. 90 Lindholm, 2012). The next section provides a literature review on sustainable UFT, stakeholder involvement and evaluation methodologies. In section 3, the MAMCA methodology is elaborated. Section 4 deals with the application of the methodology, including a case description, the stakeholders, criteria and weights. In section 5 the results are presented, followed by a discussion in section 6. Finally, the conclusions and avenues for future research are presented. 2. Literature review 2.1. Sustainable urban freight transport Based on extensive literature studies, Quak (2008) and Verlinde (2015) give an overview of the negative impact of UFT with regard to three aspects of sustainability: society, environment and economy. Improving sustainability of UFT is, amongst others, complicated because of a fragmentation of several freight flows; increased frequency, more delivery addresses and smaller volumes, driven by the growth of home deliveries, lower inventory levels and just-intime deliveries (Alho & de Abreu e Silva, 2014; Macharis & Kin, 2017). The resulting inefficiencies reveal themselves in the unnecessary presence of freight vehicles in urban areas due to low vehicle fill rates and empty running (Arvidsson, 2013). At the same time, one should be aware that UFT is highly heterogeneous and includes diverse freight flows; i.e., consumer goods to organized retail chains, to independent retailers, to individual households, but also flows of perishable goods to hotels-bars-restaurants (horeca), large flows to constructions sites and waste-related flows (Dablanc & Rodrigue, 2014). Not all flows are inefficient; organized retail is for instance often characterized by full-truckload (FTL) shipments (Quak, 2008). Local authorities are either less concerned or not aware of the interests of companies when they implement UFT measures (Ballantyne, Lindholm, & Whiteing, 2013; Lindholm, 2013; Macharis & Kin, 2017; Witkowski & Kiba-Janiak, 2014). Therefore, measures might not reach their intended effect or even produce adverse effects. Exemplary in this regard is a truck restriction policy in Manila during certain times as well as on central roads. In order to offer the same service level towards their customers, logistics service providers (LSPs) deployed more freight vehicles during a shorter period leading to more vehicle kilometres (Castro & Kuse, 2005). A study in S~ ao Paulo shows that restrictive measures

increased vehicle kilometres because of detours (Vieira & Fransoo, 2015). In Europe, several studies show that the lack of harmonization of policy measures between cities causes inefficiencies and negative environmental effects (Dablanc, Diziain, & Levifve, 2011; Quak, 2008; Russo & Comi, 2010; Van Binsbergen & Visser, 2001). Clearly, measures tailored to everyone's needs are not easily available. However, some form of harmonization has benefits, ~ uzuri particularly for shippers and LSPs (Akyol & Koster, 2013; Mun et al., 2005; Russo & Comi, 2010). Plenty of studies give an overview of different measures to regulate UFT (Anderson et al., 2005; Macharis & Kin, 2017; ~ uzuri et al., 2005; Quak, 2008). Herein a distinction between Mun two types of efforts can be made. On the one hand, there are measures that are solely the responsibility of the authorities as they are the ones governing urban areas (e.g., time windows, provision of unloading zones). On the other hand, companies are changing their UFT operations in an urban context that is regulated by authorities. Eventually, they are the ones responsible for the majority of the movements of goods (Ogden, 1992). In the latter case, the degree of involvement of the local authorities varies. Off-hour deliveries, for example, are the responsibility of a company. However, in different cities, local regulations have to permit them. Despite the type of measure and degree of involvement, consultation is stressed as essential before implementation (Bjerkan et al., 2014; € Lindholm, 2013; Osterle, Aditjandra, Vaghi, Grea, & Zunder, 2015). In recent years, more collaboration, coordination and alignment between different stakeholders in the UFT context emerged. A Freight Quality Partnerships (FQP) is an example in this respect. A FQP is used to address freight topics on a structural basis. Varying stakeholders are included (i.e., different municipal departments, environmental groups, retailers, LSPs, shippers and the local community) (Lindholm & Browne, 2013). The advantages and disadvantages of measures remain, nevertheless, dependent on the stakeholder considered as well as on the local context. The latter refers to the current infrastructure, morphology, the (freight) transport issues, and the jurisdiction and resources of the authorities (Kin, Verlinde, van Lier, & Macharis, 2016; Lindholm, 2013; ~ uzuri et al., 2005; Timms, 2014). Macharis & Melo, 2011; Mun 2.2. Stakeholders and their objectives A stakeholder is anyone who has an interest in a problem by: 1) mainly affecting it, 2) mainly being affected by it, or 3) both (Banville, Landry, Martel, & Boulaire, 1998; Macharis, 2005). Stakeholders have different interests, which are possibly conflicting. Conflicting interests between stakeholders are put forward as one of the main constraints to move to more sustainable UFT. More generally, studies in both UFT and other fields show that the difficulty, or even the lack, of stakeholder involvement in the decisionmaking process is the main shortcoming in reaching the intended goals (Beierle, 2002; Lindholm, 2013; Luyet, Schlaepfer, Parlange, & Buttler, 2012; Macharis, 2005; Macharis et al., 2012; Reed, 2008). Stakeholders mostly identified in the UFT context are the receivers, LSPs, shippers, local authorities and citizens (Behrends, 2011; Milan et al., 2015). Fig. 1 shows the different spaces where these stakeholders meet as well as their main objectives. Two remarks should be made with regard to these stakeholders. First, these groups are the most common ones, but depending on the context, other stakeholder groups might also be identified (e.g., a public transport company). Second, each stakeholder group is heterogeneous. Local authorities consist of different departments. Their interests might be affected differently by UFT. A receiver can be a big organized retail chain, the manager of a construction site but also a restaurant. The same goes for the other stakeholder groups and consequently a sub-division could be made. In section

Please cite this article in press as: Kin, B., et al., A stakeholder-based methodology to enhance the success of urban freight transport measures in a multi-level governance context, Research in Transportation Economics (2017), http://dx.doi.org/10.1016/j.retrec.2017.08.003

B. Kin et al. / Research in Transportation Economics xxx (2017) 1e14

3

Fig. 1. Relations between stakeholders in UFT context (Mommens, Kin, & Macharis, 2014; based on; Behrends, 2011).

4, the specific stakeholder groups for this case study are identified.

regard to the interests of different stakeholders prior to implementation.

2.3. Evaluation methodologies 3. Methodology: multi-actor multi-criteria analysis (MAMCA) Different methodologies are applied for ex-ante assessments of measures, policies or projects in the field of UFT. Commonly used methodologies are the (social) cost-benefit analysis ((S)CBA), costeffectiveness analysis (CEA), economic-effects analysis (EEA) and economic-impact analysis (EIA) (Browne & Ryan, 2011; Macharis et al., 2012). These methods are, nevertheless, restricted to quantified and monetized effects. Although valuation techniques exist, it remains a debate whether monetizing all effects in the context of sustainability is acceptable (e.g., human life, convenience) (Milan et al., 2015; Van Wee, 2012). In a context with different stakeholders, MCDA methods are often advocated, as it allows including both quantitative and qualitative criteria (Banville et al., 1998; Browne & Ryan, 2011; Gatta & Marcucci, 2016; Macharis, 2005; Macharis et al., 2012). In this respect, Gatta and Marcucci (2016) stress the importance of a stakeholder-specific data acquisition with regard to policy evaluation and provide an overview of different stakeholder-specific approaches in the field of UFT, including MCDA and stated preference (SP) methods. The MAMCA is an extension of traditional MCDA methods (Fandel & Spronk, 1985; Guitouni & Martel, 1998). In various cases, the MAMCA has proven to be a viable method for complex topics in which conflicting objectives of varying stakeholders are involved (Milan et al., 2015; Turcksin et al., 2011; Vermote, 2014). MCDA methods have a common value tree for all stakeholders. In a social decision context with different views, a common value tree or set of criteria for all stakeholders is not possible, as they cannot be assumed a homogeneous group (Macharis et al., 2012; Munda, 2004). The MAMCA has a value tree for each stakeholder. Therefore, different perspectives can be distinguished and a MCDA for each stakeholder group is performed (Macharis et al., 2012). Methodologically, the main difference with other MCDA methods is the way the information is brought together. Information is only aggregated in the end (output level aggregation) (Macharis et al., 2012). Herewith it allows researchers and policy-makers to evaluate different alternatives (e.g., policy measures, technologies) with

The MAMCA consists of two phases and seven steps as Fig. 2 shows. The first step is the problem definition and formulation of alternatives. As elaborated above, the problem in this study concerns the unsustainability of UFT. The local context and alternatives in the case study are described in the next section. The current situation is included as a benchmark. Hereafter, the stakeholders as well as their main objectives are identified. The objectives or criteria can be identified through a literature study and stakeholder consultation. Step three concerns the definition of criteria and weights. The criteria are weighted to indicate their relative importance to the particular stakeholder groups. The stakeholders themselves allocate weights. The pairwise comparison method of the Analytical Hierarchy Process (AHP) is mostly applied. The pairwise comparison allows indicating the relative importance of the criteria (weights) (Macharis, Springael, De Brucker, & Verbeke, 2004; Saaty, 1988). The fourth step operationalizes the criteria by constructing indicators (or metrics, or variables) to measure the impact of the alternatives. Indicators can be qualitative (e.g., service level), quantitative (e.g., CO2 emissions) and monetary. More than one indicator can be used to measure the impact on a criterion. Next, the second e synthetic or exploitation e phase follows. In this phase, it is evaluated to what extent the different alternatives contribute to the objectives of the stakeholders. The weight allocation of the stakeholders and the evaluation of the alternatives are combined. Therefore, an evaluation matrix is constructed. The evaluation of the different alternatives is done via multi-criteria analysis (MCA) methods. Different MCA techniques are able to include the multi-actor perspective, of which AHP and PROMETHEE are mostly applied (see Macharis et al. (2004) for an overview of the advantages and disadvantages of these methods). These techniques are included in the software that has recently been developed (www.mamca.be). The results (step six) give a clear overview of the advantages and disadvantages of the alternatives for each stakeholder group specifically (uni-actor) as well as for all stakeholders

Please cite this article in press as: Kin, B., et al., A stakeholder-based methodology to enhance the success of urban freight transport measures in a multi-level governance context, Research in Transportation Economics (2017), http://dx.doi.org/10.1016/j.retrec.2017.08.003

4

B. Kin et al. / Research in Transportation Economics xxx (2017) 1e14

Fig. 2. Multi-Actor multi-criteria analysis (Macharis, 2005).

combined (multi-actor). A sensitivity analysis can be applied to measure the stability of the results. The results support the decision outcome as it helps in identifying and eliminating bottlenecks for specific stakeholders. Based hereupon, alternatives can possibly be adjusted to minimize the bottlenecks for particular stakeholders. This is important for further implementation. If necessary, additional analyses can be applied on the alternatives hereafter. For instance, if a decision-maker needs to know the internal rate of a project, a CBA can be used (e.g., Kin et al., 2016). 4. Application: the Mortsel case 4.1. Problem definition and formulation of alternatives With 98% of the population living in cities, Belgium is one of the most urbanised countries in the world (United Nations, 2014). The municipality of Mortsel is located in the dense south-eastern part of the metropolitan area of Antwerp, in the Flanders region. Mortsel has a population of 25,000 and the metropolitan region holds almost 700,000 people (City of Morstel, n. d.; UNdata, 2015). The region is heavily congested with an increase in overall travel time of 26% compared to a free flow situation (TomTom, 2016). Several provincial roads cross through the centre of Mortsel. The capacity of the infrastructure is exceeded due to the large numbers of (commuting) traffic. Traffic pressure increases in case of traffic jams on the national highways. A measurement in 2011 in thirteen Flemish cities showed that Mortsel had the highest NO2 emissions in mg/m3. Traffic is responsible for 61.4% of all NO2 emissions (VMM, 2011). In addition, the municipality and more specifically the centre with the central shopping area suffers from reduced accessibility, noise nuisance and a negative impact on traffic safety. Traffic counts in the centre during rush hours showed that freight vehicles of more than 3.5 tons counted for 3e6% of all traffic (Flemish Government, n. d.). Despite this relatively low number, the impact of freight vehicles should not be underestimated for two reasons. First, the counts of freight vehicles exclude vans that are increasingly being used for deliveries (Ruesch, Schmid, Bohne, Haefeli, &

Walker, 2016). Second, counts took place during rush hours. Deliveries to retail outlets mostly happen during opening hours of shops in-between the morning and evening peak. It was also observed that freight vehicles had problems finding a vacant parking spot, thereby delaying the passing traffic. Given these problems, the infrastructure has been redeveloped in 2012 including fewer parking spots. Before 12pm, the parking spots are only available for unloading. However, enforcement appears difficult. This study solely focuses on deliveries in the city centre of Mortsel, where the majority of the shops and restaurants is located. In order to evaluate the impact of the alternatives, the number of freight vehicle movements and modal choices has to be known. Data collection is a very costly and time-consuming process. Additionally, receivers might either not be aware of the number of deliveries or not willing to share this information. Therefore, we calculated the freight vehicle movements based on assumptions from other studies. As Cherrett et al. (2012) and Alho and de Abreu e Silva (2015) show, freight vehicle movements are mostly determined by the business type in combination with the size of the sales area. This determines the freight volumes, delivery frequency and types of vehicles used per establishment. The basic reasoning is that a jeweller of the same size as a department store generates different freight volumes, has another delivery frequency and smaller freight vehicles deliver to the premises. Input for the calculations comes from these studies as well as more specific freight surveys conducted in France (Ambrosini, Patier, & Routhier, 1999), Belgium (Debauche, 2006; VIM, 2010) and the Netherlands (Buck Consultants International, 2002, 2005, 2008, 2009; Connekt, 2003). Most studies make a distinction between the modal choice delivering to each business type; a truck trailer (>18t), truck (7.5e18t), light truck (3.5e7.5t) and van/car (<3.5t). The vehicle type does not only depend on the business, but also on the fact, whether the establishment is an organized (retail) chain or not (Verlinde, 2015). In case of organized chains, modal choices are different with particularly a higher proportion of larger vehicles. For Mortsel a file with an overview of all businesses categorized

Please cite this article in press as: Kin, B., et al., A stakeholder-based methodology to enhance the success of urban freight transport measures in a multi-level governance context, Research in Transportation Economics (2017), http://dx.doi.org/10.1016/j.retrec.2017.08.003

B. Kin et al. / Research in Transportation Economics xxx (2017) 1e14

5

Fig. 3. Location of the businesses included in the study (own setup based on database provided by Municipality of Mortsel).

by NACE-code is provided. The size of the sales area is also given. Two types of businesses are excluded in this study. First, the ones not located in the main shopping area. Second, the sales area is decisive in calculating the number of freight vehicles. In the database, there are 216 businesses with a sales area of 0 m2. These are mainly service-driven companies (e.g., banks, hairdressers). Although they receive deliveries, these are often small in terms of volume as well as frequency. Businesses with 0 m2 also include horeca. Whereas horeca receives a considerable amount of deliveries, the diversity in this category is very high. It includes organized chains, small lunch shops and restaurants. This leads to unpredictability (MDS Transmodal, 2012; Tozzi, Corazza, & Musso, 2014). Above all, calculations on freight generation are simply not possible because the size of the sales area is unknown. The location of the remaining 135 businesses included in this study are shown in the Fig. 3 below. The average sales area is 108 m2 per business. Out of these 135 businesses, 40 belong to an organized chain and 95 are independent. Table 1shows that 744 freight vehicles are used to supply the retailers in the centre on a weekly basis. The delivered volumes, frequencies and modal choices are also given. Even though this seems a rather large amount, deliveries in city centres are often uncoordinated and subsequently inefficient, leading to unnecessary vehicle movements (Ljungberg & Gebresenbet, 2004). Additionally, it is not known whether a LSP delivers to one or more addresses. Therefore, it is assumed that one vehicles delivers to one address. A fictive example from the category ‘Hobby, Toy and Game shops’ with a sales area of 60 m2 is used to illustrate how freight vehicle movements are calculated. The sales volume per m2 is 0.062 m3, the received volume per week is (60
0.062) 3.72 m3; the frequency of deliveries per week for this category is 6.6 and

based on the modal choice, this volume is delivered by 0.41 trailers, 1.84 trucks, 2.25 light trucks, and 2.08 vans. In total, there are six businesses in this category leading to 39 freight vehicles per week (rounded): 2 trailers, 11 trucks, 14 light trucks and 12 vans. Based on the literature study, a list with alternatives to mitigate the UFT issues in Mortsel has been selected and discussed with the municipality and the Flemish government. Three factors contributed to the formulation of the alternatives. First, the municipality has no competence over the provincial roads but is responsible for the sidewalks and parking. Therefore, some alternatives such as limiting access to the area have been ruled out. Second, the alternatives have been formulated as realistic as possible; i.e., no measures that might lead to excessive costs for companies or the authorities. Third, the majority of the traffic problems in Mortsel is caused by cars. The alternatives, however, explicitly focus on UFT. Four alternatives are formulated: 1) Business as usual (BAU): the current situation characterized by the UFT issues described above; high numbers of transit traffic, lack of vacant (un)loading spots and therefrom arising negative externalities. 2) Urban consolidation centre (UCC): Mortsel as a community is too small to have its own UCC. There are two UCCs in Antwerp (located approximately 11 and 15 km from the centre of Mortsel) (Kin et al., 2016). Outsourcing to these UCCs is assumed. The use of a UCC is especially interesting for LSPs with small and fragmented volumes only delivering to one independent retailer in Mortsel. A fee is paid to the UCC operator. It is assumed that this leads to a positive trade-off because of saved (fuel) costs and time. Otherwise, a UCC is not used. Different studies show the willingness of LSPs to use a UCC; 13% according to Marcucci and

Table 1 Volume, frequency and modal choice per week in the centre of Mortsel. Business

Volume (in m3)

Frequency

# Trailer

# Truck

# Light truck

# Van/Car

Organized (40) Independent (95) Total (135)

596 624 1221

231 513 744

23 32 56

70 143 214

71 175 246

66 162 228

Please cite this article in press as: Kin, B., et al., A stakeholder-based methodology to enhance the success of urban freight transport measures in a multi-level governance context, Research in Transportation Economics (2017), http://dx.doi.org/10.1016/j.retrec.2017.08.003

6

B. Kin et al. / Research in Transportation Economics xxx (2017) 1e14

Danielis (2007) and 16e18% based on a review of several studies Janjevic et al. (2016) conducted. When the authorities (partly) subsidize the use, it increases to 29% (Marcucci & Danielis, 2007). In this alternative 13% e to avoid overestimation e of all vehicles delivering to independent retailers in a week (513), outsource their last mile deliveries to the UCC. This leads to a reduction by 67 freight vehicles, proportionally divided over the different types. Instead, one UCC truck delivers to the retailers every day, leading to a total number of 684 freight vehicles per week. 3) Off-hour deliveries: deliveries are allowed outside regular opening hours (early morning, 6e7am; late evening 7e11pm). Only deliveries to organized retailers are included in this alternative. Based on a pilot in Brussels, 26% of the deliveries were shifted to the off-hours (STRAIGHTSOL, 2014). There is no reduction in the number of freight vehicles but 60 of them shift to off-hours. Investments in silent material are required. 4) Loading and unloading: (un)loading is only allowed between 7am and 12pm. In the afternoon, no parking is allowed. The goal is to make the shopping environment more attractive. Traffic signs have to be adapted and enforcement is intensified. The additional costs of enforcement are offset by the additional income from fines. Because of strict enforcement, LSPs can find a vacant spot more easily and the passing traffic will be hindered less. There are no changes in the number of freight vehicle movements.

4.2. Stakeholder analysis, criteria and weights The identified stakeholders for the centre of Mortsel are in line with those in other UFT studies. The receivers are the retailers e organized and independent e located in the centre. Shippers send products to the retailers. They can be from the same company (e.g., modern retailer) or from another company. The LSPs are contracted by the shipper to deliver to the retailers. Furthermore, the municipality and the citizens are stakeholders. Due to multi-level governance, the Flemish government is included as a sixth stakeholder. The stakeholder criteria are selected through a literature study (Behrends, 2011; Melo & Costa, 2011; Milan et al., 2015; STRAIGHTSOL, 2012a). The criteria for the authorities have been discussed with them. Between 1 and 9 February 2016, 60 different stakeholders from the six groups received an email with a request to participate in the study. With the aid of a recently developed website (www.mamca.be), the problem and alternatives were introduced, followed by the request to weigh the criteria through pairwise comparisons. The Table 2 below shows the criteria and the weights allocated by respondents. The scores per stakeholder sum up to 100%. The number of consulted stakeholders and the ones who responded are given in the column on the left. Based on the input from the stakeholders, it becomes clear to what objectives they attach most importance. The weights allocated by the LSPs are relatively proportionally divided over the criteria. The shippers attach by far most importance to their customers and least to green concerns. As one can see, the response rate for the receivers was very low. Contrary to other studies, green concerns have the highest score. In order to overcome potential bias, a sensitivity analysis is conducted. A high weight for low costs for deliveries is in line with other studies. For the citizens, traffic safety, accessibility and air quality are the most important criteria. The municipality attaches most importance to good air quality and traffic safety, followed by a positive business climate. Finally, the Flemish Government gives the highest scores to traffic safety and their principle on infrastructural design (‘STOP’). This principle states that pedestrians should get priority before cyclists, public

transport and cars. A good traffic flows has the lowest score but it can be said that this is already included in the principle. 4.3. Criteria, indicators and measurement methods The indicators to measure the impact on the criteria are selected from literature (De Langhe, Gevaers, Sys, & Vanelslander, 2013; Haghshenas & Vaziri, 2012; Melo & Costa, 2011; Pissourios, 2013; STRAIGHTSOL, 2012b; Vermote, 2014). Each criterion has one or more indicators. The indicators and references are shown in the appendix. Three different ways are used to conduct the evaluation. If possible, the impact on a criterion is calculated based on the number and type of freight vehicles. This refers, for instance, to the criterion air quality, which is composed of different types of emissions. The calculations are based on the vehicle kilometres in combination with the vehicle type as well as the period of the day the vehicle is driving. The latter is important because of the flow; during more congested rush hours, vehicles emit more due to accelerating and braking (den Boer, Otten, & van Essen, 2011; Korzhenevych et al., 2014). Second, other criteria are evaluated qualitatively based on literature. For example, employee satisfaction for the LSPs scores slightly higher in the loading and unloading alternative than in the current situation. Different studies show that there is less stress for drivers when they can find a vacant unloading spot more easily (Quak, 2008; Verlinde, 2015). Finally, some criteria are calculated based on the average of other scores, as it is very complex to obtain data for them. To illustrate, the criterion green concerns is based on the scores of the criteria good air quality and minimal greenhouse gas emissions for that particular alternative. Another example is the criterion high-level service of the LSP. The LSP is contracted by the shipper to deliver to the shipper's customer, a receiver (Verlinde, 2015). Herein a certain service-level has to be provided (e.g., on time). The scores of the alternatives for high-level service are therefore based on the combination of the scores of the shippers as clients and the receivers as customers. 4.4. Overall analysis For the overall analysis (step 5), AHP is used to evaluate the impact of the alternatives on each criterion by a pairwise comparison (Macharis et al., 2004; Saaty, 1988). Scores are given by experts on a scale of 9 as shown in the Fig. 4 below. Based on the indicators, it is evaluated to what extent an alternative contributes more to a certain criterion. When both alternatives equally contribute to the criterion, ‘1’ is selected. The table in the appendix shows the data and sources used to conduct the evaluation. 5. Results 5.1. Multi- and uni-actor results The Fig. 5 below shows the multi-actor results. The aggregated scores of the weights and evaluation are shown on the y-axis (based on AHP eigenvalues method, see Saaty (1988)). The coloured lines represent the alternatives and show to what extent each one contributes to the criteria of every stakeholder (x-axis). The UCC clearly contributes the most to the criteria of all stakeholders. It is, in other words, an improvement compared to the current situation. This is especially the case for citizens. The UCC is followed by offhour deliveries. For the LSPs, both alternatives contribute in more or less the same way to their criteria. For all stakeholders there is barely a difference between the current situation and the loading and unloading alternative. This alternative contributes less to the criteria of the municipality than the current situation. By looking at the uni-actor results, the reasoning behind the

Please cite this article in press as: Kin, B., et al., A stakeholder-based methodology to enhance the success of urban freight transport measures in a multi-level governance context, Research in Transportation Economics (2017), http://dx.doi.org/10.1016/j.retrec.2017.08.003

B. Kin et al. / Research in Transportation Economics xxx (2017) 1e14

7

Table 2 Criteria, definition and weights. Stakeholder

Criterion

Definition

Weight

Logistics Service Providers (n ¼ 3/10)

Maximizing profitability

Maximum positive difference between income and variable costs for logistics' services Receiver and shipper satisfaction A positive expected return on investment Employees are satisfied with their work and working environment Positive attitude towards environmental impact Low costs for transport Satisfied retailers Pick-ups that do not disturb the operations of the shipper Positive attitude towards environmental impact Low costs for receiving goods Deliveries that do not disturb the operators of the retailer Attractive shopping environment for customers Positive attitude towards environmental impact Reduction in air polluting emissions (NOx, SO2, PM) Minimization noise nuisance Attractive shopping environment for consumers Less congestion Reduction in victims of traffic accidents with specific attention for improved safety for vulnerable road users (pedestrians, cyclists, motorcyclists, children and seniors) Positive business climate for retailers Reduction in air polluting emissions (NOx, SO2, PM) Easiness of compliance of measures Low costs for measures/enforcement Less congestion Measures are supported by citizens Minimization noise nuisance Reduction in victims of traffic accidents with specific attention for improved safety for vulnerable road users (pedestrians, cyclists, motorcyclists, children and seniors) Reduction in number of victims of traffic accidents with specific attention for improved safety for vulnerable road users (pedestrians, cyclists, motorcyclists, children and seniors) Positive impact on traffic flow Easiness of compliance of measures ‘STOP’-principle guiding (priority for pedestrians, followed by cyclists, public transport and cars) Minimization noise nuisance Reduction in air polluting emissions (NOx, SO2, PM) Reduction in CO2 emissions

17,33%

Shippers (n ¼ 3/6)

Receivers (n ¼ 2/34)

Citizens (n ¼ 4/4)

High-level service Viability of investment Employee satisfaction Green concerns Low costs for transport Satisfied customers Qualitative pick-ups Green concerns Low costs for receiving goods Convenient high-level deliveries Attractive shopping environment Green concerns Good air quality Minimal noise nuisance Attractive shopping environment Good accessibility High traffic safety

Municipality (n ¼ 3/3)

Positive business climate Good air quality Feasible enforcement Minimal cost of measures Good accessibility Social-political acceptance Minimal noise nuisance High traffic safety

Flemish Government (n ¼ 2/3)

High traffic safety

Fluent traffic flow Feasible enforcement Priority ‘STOP’-principle Minimal noise nuisance Good air quality Minimal greenhouse gas emissions

results per stakeholder group become clear. Fig. 6 shows the uniactor results of the LSPs. The bars indicate the criteria weights that correspond to the y-axis on the left. The lines represent the alternatives and the position of the lines correspond to the evaluation score shown on the y-axis on the right. The weights and evaluation of alternatives are represented aggregated on the vertical red line, which correspond to the score as shown in Fig. 5 for that particular stakeholder. As elaborated in paragraph 4.3, the evaluation is conducted in three different ways. The table in the appendix shows the justification for the evaluation of each criterion. For the LSP, off-hour deliveries (for the ones delivering to organized businesses) and the UCC (for deliveries to independent

28,72% 18,21% 22,44% 13,30% 11,66% 59,29% 20,55% 8,50% 23,75% 14,37% 27,51% 34,38% 24,74% 9,27% 11,59% 26,06% 28,34%

14,41% 19,45% 11,94% 8,25% 9,25% 7,87% 9,80% 19,03%

29,13%

2,71% 3,37% 25,41% 11,17% 15,41% 12,80%

businesses) contribute considerably to their criteria. Off-hour deliveries on a large scale can lead to considerable savings in operational costs due to lower fuel consumption and fewer delays. This explains the high score for maximizing profitability. In line with that, there is a rate of return on investments in silent material, leading to the high score for viability of investment. With regard to this criterion, the other alternatives have the same score as the current situation since no investments are required by the LSPs. The UCC also scores positively regarding green concerns, because consolidation leads to fewer vehicles and consequently to a reduction in emissions. The same applies to a smaller extent to offhour deliveries since vehicles run stationary less often outside rush hours. Regarding the loading and unloading alternative, vehicles

Fig. 4. Screenshot pairwise comparison evaluation (www.mamca.be).

Please cite this article in press as: Kin, B., et al., A stakeholder-based methodology to enhance the success of urban freight transport measures in a multi-level governance context, Research in Transportation Economics (2017), http://dx.doi.org/10.1016/j.retrec.2017.08.003

8

B. Kin et al. / Research in Transportation Economics xxx (2017) 1e14

Fig. 5. Results - Multi-actor view.

can find a vacant spot more easily. Employee satisfaction has a high score for the UCC since drivers experience less stress of having to go into a congested area. The scores of the alternatives for high-level service is based on the combination of the scores of the shippers as clients and the receivers as customers. In line with the other stakeholders, the UCC contributes the most to the criteria of the receivers (Fig. 7). The UCC slightly contributes more to high-level deliveries since LSPs only outsource deliveries when the same or a better service level will be offered to the receivers. Off-hour deliveries can have a negative impact on this criterion because drivers might put products in the wrong place. This is, however, off-set by the fact that neither the customers are disturbed by deliveries during shopping nor the receiver during opening hours. The off-hour delivery alternative has a lower score for low costs for receiving goods compared to the other alternatives because receivers e organized businesses in this case e have to invest in lockers and/or covered (un)loading docks(see Fig. 8). The UCC has the highest score on all criteria of the citizens, as

this is the only alternative actually leading to a reduction in the number of vehicles. Consequently, this affects the emissions, accessibility, safety and noise. The emissions also slightly reduce in case of off-hour deliveries because there are fewer vehicles during rush hours. Loading and unloading contributes the least to the criteria of the municipality. The UCC and the current situation contribute in the same way to the criterion minimal cost of measures. In both cases, no investments or other costs are required by the municipality. Most costs apply to the enforcement of loading and unloading. The minor cost for off-hour deliveries applies to the slight personnel costs because of adapting regulations. The same reasoning applies to the scores of the alternatives with regard to feasible enforcement. The scores for a positive business climate are based on the average scores of the receivers. Social-political acceptance applies to support by the citizens and is based on their average scores for each alternative. Priority ‘STOP’-principle and high traffic safety are the most

Fig. 6. Uni-actor results LSP.

Please cite this article in press as: Kin, B., et al., A stakeholder-based methodology to enhance the success of urban freight transport measures in a multi-level governance context, Research in Transportation Economics (2017), http://dx.doi.org/10.1016/j.retrec.2017.08.003

B. Kin et al. / Research in Transportation Economics xxx (2017) 1e14

9

Fig. 7. Uni-actor results Receiver.

important criteria for the Flemish Government. The score for the former criterion is equal for all alternatives. It has already been prioritized after the redevelopment in 2012. The scores for the other criteria are similar to those of the municipality. The UCC and off-hour deliveries both lead to fewer vehicle movements during busy hours; finding a vacant spot more easily contributes significantly to the criterion on fluent traffic flow. 5.2. Sensitivity analyses Sensitivity analyses are conducted to test the robustness of the results obtained. The purpose is to check the stability of the results, or impact of the allocated weights. This is important for the implementation phase. Because weights are only allocated by a limited number of respondents from a stakeholder group, results can differ when other or more respondents allocate weights (Macharis et al., 2012; STRAIGHTSOL, 2014). Therefore, the weights of the criteria are changed from low (0%) to high (100%) to see whether the ranking of the results remain the same or not. The software is used to change the weights of each criterion per stakeholder. This immediately shows whether it leads to a change in the ranking of the results. The numbers show, below (<) or above (>) what weight, a rank reversal of the alternatives occurs. The results for the LSPs are relatively robust. When the weight for the criterion maximizing profitability increases (>28%), off-hour deliveries and the UCC change in the ranking. The same happens when the weights for the criteria green concerns (<4%) and employee satisfaction (<3%) become lower. The scores for the other two alternatives are barely affected. A similar change takes place for the shippers when the weight for low costs for transport increases (>40%). The weights for the other criteria are robust. As mentioned, the response for the receivers was low with a relatively high weight for green concerns. However, when the weight of this criterion changes from 34% to 0% thereby increasing the weights of the other criteria proportionally, the ranking remains the same. The same goes for the other criteria. The results for this group as well as for the citizens are thus very robust. The ranking for the municipality only changes when the criterion minimal cost of measures increases (>30%). In this case, the score of BAU increases vis- a-vis offhour deliveries. Finally, the ranking for the Flemish Government

remains robust, except the weight for feasible enforcement increases (>27%). 6. Discussion 6.1. Alternatives The three alternatives are separately evaluated. In practice they can also be implemented simultaneously since they are not conflicting. The LSPs that use a UCC are considered others than those delivering during off-hours. Vacant (un)loading zones apply to all LSPs. Although not evaluated, the expectation is that the difference between a combined alternative and the current situation widens positively. Although the results show that improvements to the current situation are possible, it does not mean that changes are actually going to take place. To stimulate this, awareness has to be raised. A possibility in this regard is a FQP that can be organized on a local as well as a regional level. The latter is particularly interesting since the case under study is characterized by multi-level governance. Additionally, Mortsel is located in the metropolitan region of Antwerp and borders several other municipalities. Harmonization of measures is desirable. The regional government can be facilitating herein. An interesting alternative could be a UCC serving multiple smaller municipalities. A structural platform on a regional scale also seems interesting for off-hour deliveries. Offhour deliveries only become interesting for companies when they can be conducted to deliver to multiple addresses in a region. An interesting addition to the alternative on loading and unloading would be to use sensors for easier enforcement. Sensors have been applied to monitor vacant spots for loading and unloading in Lisbon. The results of the MAMCA in Lisbon showed that sensors contributed positively to the criteria of the different stakeholders (STRAIGHTSOL, 2014). A limitation of this study is the exclusion of several freight flows such as horeca, service trips and home deliveries. Deliveries to horeca are heterogeneous, but have several characteristics that complicate it to include them into the alternatives; i.e., perishability and the time of deliveries (Danielis, Rotaris, & Marcucci, 2010; Marcucci & Danielis, 2007). Wholesalers delivering to horeca are often relatively efficiently loaded (Danielis et al., 2010).

Please cite this article in press as: Kin, B., et al., A stakeholder-based methodology to enhance the success of urban freight transport measures in a multi-level governance context, Research in Transportation Economics (2017), http://dx.doi.org/10.1016/j.retrec.2017.08.003

10

B. Kin et al. / Research in Transportation Economics xxx (2017) 1e14

Fig. 8. Uni-actor results Municipality of Mortsel.

Contrariwise, owners of small businesses might go to the wholesaler with their own vehicle. A study in Nijmegen, the Netherlands, shows that these vehicles have a fill rate of less than 25% (Buck Consultants International, 2005). This is, nevertheless, difficult to regulate. Service-oriented businesses might even be more diverse. In several cases, they receive office deliveries that are very suitable to go through a UCC (Browne, Allen, & Leonardi, 2011). For businesses with small volumes (e.g., hairdresser) or documents (e.g., banks), bike couriers are an interesting option (Gruber, Kihm, & Lenz, 2014). Finally, home deliveries, which also affect residential neighbourhoods, are expected to increase in the future (Visser, Nemoto, & Browne, 2014). The use of a UCC to bundle small volumes is a very promising option for deliveries to individual households, as is the use of manned pick-up points (e.g, Kiala) and lockers (Morganti, Dablanc, & Fortin, 2014). In this respect, the emergence of omni-channel retail should be closely taken into account. Omni-channel allows the consumer to use different channels (physical, online) interchangeably. One of the effects are shipments by retailers (Verhoef, Kannan, & Inman, 2015). 6.2. Application MAMCA Depending on the context and the purpose of the study, different methods for ex-ante assessments are available (see section 2.3). In this case, the MAMCA has been determined as the most appropriate method. First, a MCDA is more applicable than quantitative methods, because not all effects can be quantified or monetized. Second, compared to a traditional MCDA that has a common value tree for all stakeholders, it allows incorporating the different interests of varying stakeholders. Herewith it provides an integrated evaluation. Decision-makers are supported in their decision-making process as well as in the decision outcome. In this respect, it provides a tool for participatory decision-making. Moreover, by incorporating the views of different stakeholders, a

common understanding is created (Macharis et al., 2012; Ward et al., 2016). Vermote (2014) adds a more democratic representation and social learning. The results allow identifying potential bottlenecks of an alternative. Based hereupon, the decision-maker can try to minimize these in order to enhance the success of future implementation. Therefore, one can look at the uni-actor results. When an alternative negatively contributes to the criteria of a certain stakeholder group, the uni-actor results show to what specific criteria this alternative contributes the least. Herewith, the alternative can be adapted before future implementation in such a way that it minimizes the negative impact to the extent possible. If necessary, other analyses such as an EEA can be applied on a (modified) alternative. Compared to similar UFT studies, this case is characterized by multi-level governance, with the regional and local authorities having different jurisdictions. Extending upon this, Te Boveldt et al. (2016) propose a competence-based multi-criteria analysis (COMCA). The COMCA addresses the challenge of interjurisdictional decision-making and makes it possible to incorporate hierarchical distinctions between stakeholders, based on their role in a specific project. In this study, stakeholder input on the weights is acquired through an online survey. Workshops are used in other cases. The benefit of a workshop is that stakeholders can be informed more clearly on the purpose of the study and the alternatives. The drawback is that certain stakeholder groups e particularly companies e have a low attendance (Lebeau, 2016). Whether a workshop is organized, or participation through an online survey is requested, there is always the risk of bias (Beierle, 2002). Certain groups might be underrepresented. Above all, within stakeholder groups, there can be significant differences, and the participating respondents might not be representative for the whole group (Lebeau, 2016; Vermote, 2014). A sensitivity analysis can minimize this risk to some extent. Data collection for the evaluation is a very time-consuming, complex and expensive process. In this study,

Please cite this article in press as: Kin, B., et al., A stakeholder-based methodology to enhance the success of urban freight transport measures in a multi-level governance context, Research in Transportation Economics (2017), http://dx.doi.org/10.1016/j.retrec.2017.08.003

B. Kin et al. / Research in Transportation Economics xxx (2017) 1e14

therefore, the evaluation to be done by assumptions and calculations based on literature and other case studies. The evaluation itself is done by experts. Another possibility is stakeholder input based on judgments (Macharis et al., 2012).

7. Conclusion The current supply of goods to retailers in Mortsel causes different challenges. Improving the sustainability through measures is complicated because of the multi-level governance context with different jurisdictions over infrastructure. Consequently, a trade-off between authorities is required and some potential measures are excluded (e.g., regulating access to the area with a low emission zone). Three alternatives for delivering shops in the centre of Mortsel have been formulated and evaluated. The results show that all alternatives contribute to the criteria of the different stakeholders involved. Per alternative, the advantages and disadvantages for every stakeholder are identified. By minimizing potential bottlenecks early in the process, alternatives can be modified and made more acceptable for specific stakeholders. This enhances the successful and acceptable implementation of measures. The UCC clearly contributes the most to the criteria of the stakeholders, followed by off-hour deliveries. Loading and unloading is barely an improvement to the current situation. The alternatives are not conflicting and can therefore be implemented simultaneously in order to enhance their positive impact as shown by the evaluation. Contrary to similar studies, it can be said that there is a high probability for consensus between the stakeholders; no alternative has a high score for one group and a low one for another group. In line with this, the results show no conflict between the local and regional authorities. Despite the results, implementation of the proposed measures does not automatically take place. The MAMCA already brings the stakeholders and their interests together. Hereafter, this can be more formalized in a platform such as a FQP. The location of Mortsel in a larger metropolitan area necessitates involving stakeholders from neighbouring municipalities to tackle the challenges and harmonize measures. The regional authorities could play a mediating role between different municipalities that are all affected by the same vehicle movements and policies. For future implementation, attention also has to be paid to a more inclusive policy towards UFT also taking

11

into account flows such as omni-channel, construction, waste and office deliveries. The MAMCA has proven to be a useful evaluation framework concerning a complex decision problem in which several stakeholders have different, possibly conflicting, objectives. Stakeholders are actively involved in the evaluation by allocating weights to their criteria. This already raises awareness. In general, a complex topic such as UFT is presented more concretely, which contributes to a grounded and more democratic decision-making process. Instead of workshops, the weights have been collected online via a website (www.mamca.be). This allows immediate aggregation of the weights with the evaluation of the alternatives. For the evaluation, assumptions on the freight vehicle movements had to be made. Since (freight) traffic counts are a very costly and time-consuming process, more focus could be placed on a simulation model that takes recipient heterogeneity and vehicle types into account. Big data analytics, to more specifically monitor freight vehicle movements, could be explored. The diversity within stakeholder groups, such as different governmental departments or different types of receivers, has to be taken into account. With the MAMCA, this can be solved by either splitting the stakeholder groups or making subgroups. Interesting avenues to explore are how to actively involve the stakeholders in the development of alternatives, and collect large-scale input from specific stakeholder groups, particularly residents. Future work with the COMCA also focuses on interjurisdictional decision-making whereby a distinction between institutional stakeholders is made, based on their competences.

Acknowledgement This research was supported by the Research Centre on Commodity and Passenger Flows (MOBILO). We would like to thank the City of Mortsel for their cooperation. We also express our gratitude to the anonymous reviewers who contributed to improving this work.

Appendix A The table below shows the evaluation for the UCC (1), off-hour deliveries (2) and loading and unloading (3).

Table 3 Indicators and assumptions evaluation. Criterion

Indicator

Elaboration evaluation

Source

1.1 Maximizing profitability

Operational benefits

LSPs only outsource when it leads to a benefit or break-even (1). 11.7% decrease of costs (2). Slight decrease in unloading time (3). Average 2.1e3.4

No investment required (1,3). Ratio of 7% (2)

1.4 Employee satisfaction

Ratio gained or lost on total investment. Employee satisfaction

Lebeau, 2016; STRAIGHTSOL, 2014; Verlinde, 2015 Lebeau, 2016; STRAIGHTSOL, 2014; Verlinde, 2015 STRAIGHTSOL, 2014; Verlinde, 2015 Lebeau, 2016; Verlinde, 2015

1.5 Green concerns 2.1 Low costs for transport

Costs per delivered item

1.2 High-level service

1.3 Viability of investment

2.2 Satisfied customers 2.3 Qualitative pick-ups 2.4 Green concerns 3.1 Low costs for receiving goods

Pick-up on agreed time Costs per received item

Driver does not have to enter the city, but still faces congestion on highways and rush hours (1), no congestion but working during the night (2), less stress when looking for a vacant unloading spot (3) Score indicators 6.6 and 6.7 (50/50) Shipper does not have influence, responsibility LSP (1), lower costs because same company (2), no impact (3) Receiver are customer of shipper; average score of receivers (3.1e3.4) No influence on pick-ups Score indicators 6.6 and 6.7 (50/50) Receivers do not pay for the UCC (1), costs for organized chain (2), no difference for receiver (3)

STRAIGHTSOL, 2014 STRAIGHTSOL, 2014

VIL, 2016 STRAIGHTSOL, 2014 Lebeau, 2016; STRAIGHTSOL, 2014 (continued on next page)

Please cite this article in press as: Kin, B., et al., A stakeholder-based methodology to enhance the success of urban freight transport measures in a multi-level governance context, Research in Transportation Economics (2017), http://dx.doi.org/10.1016/j.retrec.2017.08.003

12

B. Kin et al. / Research in Transportation Economics xxx (2017) 1e14

Table 3 (continued ) Criterion

Indicator

3.2 Convenient high-level deliveries 3.3 Attractive shopping environment

3.4 Green concerns 4.1 Good air quality

4.2 Minimal noise nuisance

Reduction Reduction Reduction Reduction

NOx (g/vkm) SO2 (g/vkm) PM 2.5 (g/vkm) noise level

4.3 Attractive shopping environment

4.4 Good accessibility

Number of freight vehicles

4.5 High traffic safety 5.1 Positive business climate 5.2 Good air quality

Reduction NOx (g/vkm) Reduction SO2 (g/vkm) Reduction PM 2.5 (g/vkm)

5.3 Feasible enforcement

Elaboration evaluation

Source

Same service otherwise not outsourced (1), driver might put products in wrong place, but products are delivered earlier and outside shopping hours (2), no difference (3) Assumption that shopping environment becomes more attractive when there is a good air quality, minimal noise nuisance, less congestion and higher traffic safety; score indicators 4.1, 4.2, 4.4, 4.5 (25/25/25/25) Score indicators 6.6 and 6.7 (50/50) Calculated based on vehicle type and period of the day driving

STRAIGHTSOL, 2014

Calculated with marginal external costs of noise, based on vehicle types and period of the day driving Assumption that shopping environment becomes more attractive when there is a good air quality, minimal noise nuisance, less congestion and higher traffic safety; score indicators 4.1, 4.2, 4.4, 4.5 (25/25/25/25) Calculated based on number of vehicles; vehicles disappear from rush hours (1,2), minimal improvement because drivers can find a vacant spot more easily (3) Improvement because fewer vehicles (1), fewer vehicles during rush hours but still during off-hours (2), less maneuvers (3) Average score receivers (3.1e3.4) Calculated based on vehicle type and period of the day driving

Korzhenevych et al., 2014; Van Lier, 2014

No enforcement required (1), adaptation regulation (2) strict enforcement required (3)

~ uzuri et al., 2005; Mun Quak, 2008; STAIGHTSOL, 2014 ~ uzuri et al., 2005; Mun Quak, 2008; STAIGHTSOL, 2014

5.4 Minimal cost of measures

Costs for introducing measures by local authorities

No costs (1), adapt regulation leads to minimal personnel cost (2), additional costs (3)

5.5 Good accessibility

Number of freight vehicles

5.6 Social-political acceptance 5.7 Minimal noise nuisance

Reduction noise level

Calculated based on number of vehicles; vehicles disappear from rush hours (1,2), minimal improvement because drivers can find a vacant spot more easily (3) Average score citizens (4.1e4.5) Calculated with marginal external costs of noise, based on vehicle types and period of the day driving Improvement because fewer vehicles (1), fewer vehicles during rush hours but still during off-hours (2), less maneuvers (3) Idem Calculated based on number of vehicles; vehicles disappear from rush hours (1,2), minimal improvement because drivers can find a vacant spot more easily (3) Flemish Government is not responsbile for enforcement (1,2,3) Since redevelopment infrastructure, emphasis is already on ‘STOP’ (1,2,3) Calculated with marginal external costs of noise, based on vehicle types and period of the day driving Calculated based on vehicle type and period of the day driving

5.8 High traffic safety 6.1 High traffic safety 6.2 Fluent traffic flow

Number of freight vehicles

6.3 Feasible enforcement 6.4 Priority ‘STOP’-principle 6.5 Minimal noise nuisance

Reduction noise level

6.6 Minimal greenhouse gas emissions 6.7 Good air quality

Reduction CO2 (g/vkm) Reduction NOx (g/vkm) Reduction SO2 (g/vkm) Reduction PM 2.5 (g/vkm)

Calculated based on vehicle type and period of the day driving

References Akyol, D., & Koster, R. (2013). Non-dominated time-window policies in city distribution. Production and Operations Management, 22(3), 739e751. Alho, A. R., & de Abreu e Silva, J. (2014). Analyzing the relation between land-use/ urban freight operations and the need for dedicated infrastructure/enforcement - application to the city of Lisbon. Research in Transportation Business and Management, 11, 85e97. http://doi.org/10.1016/j.rtbm.2014.05.002. Alho, A. R., & de Abreu e Silva, J. (2015). Lisbon's establishment-based freight survey: Revealing retail establishments' characteristics, goods ordering and delivery processes. European Transport Research Review, 7(2). http://doi.org/10. 1007/s12544-015-0163-7. Ambrosini, C., Patier, D., & Routhier, J.-L. (1999). R esultats de l’enquete quantitative  Marseille. r ealis ee a Anderson, S., Allen, J., & Browne, M. (2005). Urban logisticseehow can it meet policy makers' sustainability objectives? Journal of Transport Geography, 13(1), 71e81. http://doi.org/10.1016/j.jtrangeo.2004.11.002. Arvidsson, N. (2013). The milk run revisited: A load factor paradox with economic and environmental implications for urban freight transport. Transportation

STRAIGHTSOL, 2014 den Boer et al., 2011; Korzhenevych et al., 2014

Korzhenevych et al., 2014; STRAIGHTSOL, 2014 den Boer et al., 2011; Korzhenevych et al., 2014

Korzhenevych et al., 2014; Van Lier, 2014 Korzhenevych et al., 2014; STRAIGHTSOL, 2014 Idem

Korzhenevych et al., 2014; Van Lier, 2014 den Boer et al., 2011; Korzhenevych et al., 2014 den Boer et al., 2011; Korzhenevych et al., 2014

Research Part A: Policy and Practice, 51, 56e62. http://doi.org/10.1016/j.tra.2013. 04.001. Ballantyne, E. E. F., Lindholm, M., & Whiteing, A. (2013). A comparative study of urban freight transport planning: Addressing stakeholder needs. Journal of Transport Geography, 32, 93e101. http://doi.org/10.1016/j.jtrangeo.2013.08.013. Banville, C., Landry, M., Martel, J., & Boulaire, C. (1998). A stakeholder approach to MCDA. Systems Research, 15, 15e32. Behrends, S. (2011). Urban freight transport sustainability. The interaction of urban freight and intermodal transport. Chalmers University of Technology. Beierle, T. C. (2002). The quality of stakeholder-based decisions. Risk Analysis, 22(4), 739e749. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/12224747. Bjerkan, K. Y., Sund, A. B., & Nordtømme, M. E. (2014). Stakeholder responses to measures green and efficient urban freight. Research in Transportation Business & Management, 11, 32e42. http://doi.org/10.1016/j.rtbm.2014.05.001. den Boer, E., Otten, M., & van Essen, H. (2011). STREAM International Freight 2011. Comparison of various transport modes on a EU scale with the STREAM database. Delft. Browne, M., Allen, J., & Leonardi, J. (2011). Evaluating the use of an urban consolidation centre and electric vehicles in central London. IATSS Research, 35(1), 1e6. http://doi.org/10.1016/j.iatssr.2011.06.002.

Please cite this article in press as: Kin, B., et al., A stakeholder-based methodology to enhance the success of urban freight transport measures in a multi-level governance context, Research in Transportation Economics (2017), http://dx.doi.org/10.1016/j.retrec.2017.08.003

B. Kin et al. / Research in Transportation Economics xxx (2017) 1e14 Browne, D., & Ryan, L. (2011). Comparative analysis of evaluation techniques for transport policies. Environmental Impact Assessment Review, 31(3), 226e233. http://doi.org/10.1016/j.eiar.2010.11.001. Buck Consultants International. (2002). Flexibel goederenvervoer. Den Haag. Buck Consultants International. (2005). 0-meting vervoersbewegingen binnenstad nijmegen. Buck Consultants International. (2008). Bevoorradingsonderzoek binnenstad breda. Buck Consultants International. (2009). Bevoorradingsprofiel utrecht. Castro, J. T., & Kuse, H. (2005). Impacts of large truck restrictions in freight carrier operations in Metro Manila. Journal of the Eastern Asia Society for Transportation Studies, 6, 2947e2962. Cherrett, T., Allen, J., McLeod, F., Maynard, S., Hickford, A., & Browne, M. (2012). Understanding urban freight activity e key issues for freight planning. Journal of Transport Geography, 24, 22e32. http://doi.org/10.1016/j.jtrangeo.2012.05. 008. Connekt. (2003). Bevoorradingsprofiel. Utrecht Binnenstad. Dablanc, L. (2007). Goods transport in large European cities: Difficult to organize, difficult to modernize. Transportation Research Part A, 41(3), 280e285. http:// doi.org/10.1016/j.tra.2006.05.005. Dablanc, L., Diziain, D., & Levifve, H. (2011). Urban freight consultations in the Paris region. European Transport Research Review, 3(1), 47e57. http://doi.org/10.1007/ s12544-011-0049-2. Dablanc, L., & Rodrigue, J.-P. (2014). City Logistics: Towards a global typology. In Transport research Arena 2014. Paris. Danielis, R., Rotaris, L., & Marcucci, E. (2010). Urban freight policies and distribution channels: A discussion based on evidence from Italian cities. European Transport/Trasporti Europei, 46, 114e146. De Langhe, K., Gevaers, R., Sys, C., & Vanelslander, T. (2013). Dataverzameling stedelijke distributie: Stedelijke indicatoren en dataverzamelmethodes (No. D/2013/ 11.528/1). ge. Debauche, W. (2006). Le transport de marchandises dans le centre ville de Lie Annexe au bulletin CRR n., 66. Fandel, G., & Spronk, J. (1985). Multiple criteria decision methods and applications. Berlin: Springer Verlag. Gatta, V., & Marcucci, E. (2016). Stakeholder-specific data acquisition and urban freight policy evaluation: Evidence, implications and new suggestions. Transport Reviews, 36(5), 585e609. http://doi.org/10.1080/01441647.2015.1126385. Gruber, J., Kihm, A., & Lenz, B. (2014). A new vehicle for urban freight? An ex-ante evaluation of electric cargo bikes in courier services. Research in Transportation Business and Management, 11, 53e62. http://doi.org/10.1016/j.rtbm.2014.03.004. Guitouni, A., & Martel, J.-M. (1998). Tentative guidelines to help choosing an appropriate MCDA method. European Journal of Operational Research, 109(2), 501e521. http://doi.org/10.1016/S0377-2217(98)00073-3. Haghshenas, H., & Vaziri, M. (2012). Urban sustainable transportation indicators for global comparison. Ecological Indicators, 15(1), 115e121. http://doi.org/10.1016/j. ecolind.2011.09.010. Janjevic, M., Lebeau, P., Ndiaye, A. B., Macharis, C., Van Mierlo, J., & Nsamzinshuti, A. (2016). Strategic scenarios for sustainable urban distribution in the Brusselscapital region using urban consolidation centres. Transportation Research Procedia, 12, 598e612. http://doi.org/10.1016/j.trpro.2016.02.014. Kin, B., Verlinde, S., van Lier, T., & Macharis, C. (2016). Is there life after subsidy for an urban consolidation centre? An investigation of the total costs and benefits of a privately- initiated concept. Transportation Research Procedia, 12, 357e369. http://doi.org/10.1016/j.trpro.2016.02.072. €cker, J., Holtkamp, M., Meier, H., Gibson, G., et al. Korzhenevych, A., Dehnen, N., Bro (2014). Update of the handbook on external costs of transport. Final Report. Lebeau, P. (2016). Towards the electrification of city logistics? Brussel: Vrije Universiteit. Lindholm, M. E. (2012). Enabling sustainable development of urban freight from a local authority perspective. Chalmers University of Technology. Lindholm, M. (2013). Urban freight transport from a local authority perspective e a literature review. European Transport/Trasporti Europei, 3(54), 1e37. Lindholm, M., & Behrends, S. (2012). Challenges in urban freight transport planning e a review in the Baltic Sea Region. Journal of Transport Geography, 22, 129e136. http://doi.org/10.1016/j.jtrangeo.2012.01.001. Lindholm, M., & Browne, M. (2013). Local authority cooperation with urban freight stakeholders: A comparison of partnership approaches. European Journal of Transport and Infrastructure Research, 13(1), 20e38. Ljungberg, D., & Gebresenbet, G. (2004). Mapping out the potential for coordinated goods distribution in city centres: The case of Uppsala. International Journal of Transport Management, 2(3e4), 161e172. http://doi.org/10.1016/j.ijtm.2005.07. 001. Luyet, V., Schlaepfer, R., Parlange, M. B., & Buttler, A. (2012). A framework to implement Stakeholder participation in environmental projects. Journal of Environmental Management, 111, 213e219. http://doi.org/10.1016/j.jenvman. 2012.06.026. Macharis, C. (2005). The importance of stakeholder analysis in freight transport. European Transport/Trasporti Europei, 25e26, 114e126. Macharis, C., & Kin, B. (2017). The 4 A's of sustainable city Distribution: Innovative solutions and challenges ahead. International Journal of Sustainable Transportation, 11(2), 59e71. http://doi.org/10.1080/15568318.2016.1196404. Macharis, C., & Melo, S. (2011). City distribution and urban freight transport: Multiple perspectives. Edward Elgar Publishing. http://doi.org/10.4337/9780857932754. 00005. Macharis, C., Springael, J., De Brucker, K., & Verbeke, A. (2004). PROMETHEE and

13

AHP: The design of operational synergies in multicriteria analysis. Strengthening PROMETHEE with ideas of AHP. European Journal of Operational Research, 153(2), 307e317. http://doi.org/10.1016/S0377-2217(03)00153-X. Macharis, C., Turcksin, L., & Lebeau, K. (2012). Multi actor multi criteria analysis (MAMCA) as a tool to support sustainable decisions: State of use. Decision Support Systems, 54(1), 610e620. http://doi.org/10.1016/j.dss.2012.08.008. Marcucci, E., & Danielis, R. (2007). The potential demand for a urban freight consolidation centre. Transportation, 35(2), 269e284. http://doi.org/10.1007/ s11116-007-9147-3. Marcucci, E., & Stathopoulos, A. (2012). Multi-level governance and transport Policy: The case of local roads in Italy. International Journal of Transport Economics, 34(1), 15e38. Marsden, G., & Rye, T. (2010). The governance of transport and climate change. Journal of Transport Geography, 18(6), 669e678. http://doi.org/10.1016/j. jtrangeo.2009.09.014. MDS Transmodal. (2012). DG MOVE European Commission: Study on urban freight transport. Melo, S., & Costa, A. (2011). Definition of a set of indicators to evaluate the performance of urban goods distribution initiatives. In C. Macharis, & S. Melo (Eds.), City distribution and urban freight Transport: Multiple perspectives (pp. 120e147). Cheltenham, UK: Edward Elgar. Milan, L., Kin, B., Verlinde, S., & Macharis, C. (2015). Multi-actor multi-criteria analysis for sustainable city distribution: A new assessment framework. International Journal of Multicriteria Decision Making, 5(4), 334e354. Mommens, K., Kin, B., & Macharis, C. (2014). Evaluatiekader voor duurzame stadsbevoorradingsconcepten. Antwerpen. Retrieved from http://www. steunpuntmobilo.be. Morganti, E., Dablanc, L., & Fortin, F. (2014). Final deliveries for online shopping: The deployment of pickup point networks in urban and suburban areas. Research in Transportation Business and Management, 11(February 2012), 23e31. http://doi. org/10.1016/j.rtbm.2014.03.002. Munda, G. (2004). Social multi-criteria evaluation: Methodological foundations and operational consequences. European Journal of Operational Research, 158(3), 662e677. http://doi.org/10.1016/S0377-2217(03)00369-2. ~ uzuri, J., Larran ~ eta, J., Onieva, L., & Corte s, P. (2005). Solutions applicable by Mun local administrations for urban logistics improvement. Cities, 22(1), 15e28. http://doi.org/10.1016/j.cities.2004.10.003. Ogden, K. W. (1992). Urban goods movement: A guide to policy and planning. Cambridge, United Kingdom: University Press. € Osterle, I., Aditjandra, P. T., Vaghi, C., Grea, G., & Zunder, T. H. (2015). The role of a structured stakeholder consultation process within the establishment of a sustainable urban supply chain. Supply Chain Management: An International Journal, 20(3), 284e299. http://doi.org/10.1108/SCM-05-2014-0149. Pissourios, I. a. (2013). An interdisciplinary study on indicators: A comparative review of quality-of-life, macroeconomic, environmental, welfare and sustainability indicators. Ecological Indicators, 34, 420e427. http://doi.org/10.1016/j. ecolind.2013.06.008. Quak, H. (2008). Sustainability of urban freight transport. Retail distribution and local regulations in cities. Erasmus University Rotterdam. Reed, M. S. (2008). Stakeholder participation for environmental management: A literature review. Biological Conservation, 141(10), 2417e2431. http://doi.org/10. 1016/j.biocon.2008.07.014. Ruesch, M., Schmid, T., Bohne, S., Haefeli, U., & Walker, D. (2016). Freight transport with vans: Developments and measures. In Transportation research procedia (Vol. 12)Elsevier B.V.. pp. 79e92 http://doi.org/10.1016/j.trpro.2016.02.049. Russo, F., & Comi, A. (2010). A classification of city logistics measures and connected impacts. Procedia - Social and Behavioral Sciences, 2(3), 6355e6365. http://doi. org/10.1016/j.sbspro.2010.04.044. Saaty, T. L. (1988). The analytical Hierarchy process. New York: McGrawHill. Stathopoulos, A., Valeri, E., & Marcucci, E. (2012). Stakeholder reactions to urban freight policy innovation. Journal of Transport Geography, 22, 34e45. http://doi. org/10.1016/j.jtrangeo.2011.11.017. STRAIGHTSOL. (2012a). Deliverable 3.2. Report on stakeholders, criteria and weights. STRAIGHTSOL. (2012b). Deliverable 3.3. Description of indicators, KPIs and measurement methods. STRAIGHTSOL. (2014). Deliverable 5.4. Final evaluation of all STRAIGHTSOL city distribution concepts by use of the MAMCA. Brussels. Te Boveldt, G., Van Raemdonck, K., & Macharis, C. (2016). A new railway tunnel under Brussels? Assessing political feasibility and desirability with Competence-based Multi Criteria Analysis. Transport Policy [submitted]. Timms, P. (2014). Transferability of urban freight transport measures: A case study of cariacica (Brazil). Research in Transportation Business & Management, 11, 63e74. http://doi.org/10.1016/j.rtbm.2014.02.001. TomTom. (2016). Tomtom traffic index. Measuring congestion worldwide. Available from: https://www.tomtom.com/nl_be/trafficindex/list [6 September, 2016]. Tozzi, M., Corazza, M. V., & Musso, A. (2014). Urban goods movements in a sensitive context: The case of Parma. Research in Transportation Business and Management, 11, 134e141. http://doi.org/10.1016/j.rtbm.2014.03.003. Turcksin, L., Macharis, C., Lebeau, K., Boureima, F., Van Mierlo, J., Bram, S., et al. (2011). A multi-actor multi-criteria framework to assess the stakeholder support for different biofuel options: The case of Belgium. Energy Policy, 39(1), 200e214. http://doi.org/10.1016/j.enpol.2010.09.033. UNdata. (2015). City population by sex, city and city type (Belgium, Antwerp, urban agglomeration). Retrieved from: http://data.un.org/Data.aspx? q¼cityþpopulation&d¼POP&f¼tableCode%3a240 [6 September, 2016].

Please cite this article in press as: Kin, B., et al., A stakeholder-based methodology to enhance the success of urban freight transport measures in a multi-level governance context, Research in Transportation Economics (2017), http://dx.doi.org/10.1016/j.retrec.2017.08.003

14

B. Kin et al. / Research in Transportation Economics xxx (2017) 1e14

United Nations. (2014). World urbanization Prospects: The 2014 revision, highlights (ST/ESA/SER.A/352). Van Binsbergen, A., & Visser, J. (2001). Innovation steps towards efficient goods distribution systems for urban areas. Efficiency improvement of goods distribution in urban areas. Technische Universiteit Delft. Van Lier, T. (2014). The development of an external cost calculator framework for evaluating the sustainability of transport solutions. Vrije Universiteit Brussel. Van Wee, B. (2012). How suitable is CBA for the ex-ante evaluation of transport projects and policies? A discussion from the perspective of ethics. Transport Policy, 19(1), 1e7. http://doi.org/10.1016/j.tranpol.2011.07.001. Verhoef, P. C., Kannan, P. K., & Inman, J. J. (2015). From multi-channel retailing to omni-channel retailing. Journal of Retailing, 91(2), 174e181. http://doi.org/10. 1016/j.jretai.2015.02.005. Verlinde, S. (2015). Promising but challenging urban freight transport solutions: Freight flow consolidation and off-hour deliveries. Vrije Universiteit Brussel; Universiteit Gent. Vermote, L. (2014). Sustainable mobility and decision-making. Participatory assessment of mobility policies in Belgium and Palestine. Vrije Universiteit Brussel. Vieira, J. G. V., & Fransoo, J. C. (2015). How logistics performance of freight operators is affected by urban freight distribution issues. Transport Policy, 44, 37e47.

http://doi.org/10.1016/j.tranpol.2015.06.007. VIL. (2016). Dal- en nachtdistributie. Available from: http://vil.be/project/dal-ennachtdistributie/ [22 March, 2016]. Visser, J., Nemoto, T., & Browne, M. (2014). Home delivery and the impacts on urban freight transport: A review. In Procedia - Social and Behavioral Sciences (Vol. 125, pp. 15e27). Elsevier B.V.. http://doi.org/10.1016/j.sbspro.2014.01.1452 Vlaams Instituut voor Mobiliteit. (2010). Eindrapport pilots d-via. Innovatieve oplossingen voor stedelijke distributie: Haalbare modellen rond vraaggestuurd bundelen en stedelijke distributie. VMM. (2011). NO2-meetcampagne met passieve samplers in steden in 2010. https:// www.vmm.be/publicaties/no2-meetcampagne-met-passieve-samplers-insteden-in-2010. (Accessed 22 October 2014). Ward, J. E., Dimitriou, H. T., & Dean, M. (2016). The application of policy-led multicriteria analysis to mega transport infrastructure project appraisal. Research in Transportation Economics, 58, 21e45. http://doi.org/10.1016/j.retrec.2016.08. 003. Witkowski, J., & Kiba-Janiak, M. (2014). The role of local governments in the development of city logistics. Procedia - Social and Behavioral Sciences, 125, 373e385. http://doi.org/10.1016/j.sbspro.2014.01.1481.

Please cite this article in press as: Kin, B., et al., A stakeholder-based methodology to enhance the success of urban freight transport measures in a multi-level governance context, Research in Transportation Economics (2017), http://dx.doi.org/10.1016/j.retrec.2017.08.003