Integrating deliveries to consumers and businesses and returns using micro- hubs in city distribution

1

Integrating deliveries to consumers and

businesses and returns using

micro-hubs in city distribution

A design science approach to create efficient and sustainable e-commerce logistics

Master Thesis

Anouk Oude Vrielink S2515229

University of Groningen Faculty of Economics and Business MSc. Technology and Operations Management

January 2017

2

A

4

5.4 Design micro-hub ... 30

5.5 Feasibility of the solution ... 31

5.6 Validity ... 33 6 Discussion ... 35 6.1 Effects on traffic ... 35 6.2 Self-collection point ... 35 6.3 Business model ... 36 6.4 Other ... 36 6.5 Redesign ... 37 7 Conclusion ... 39 7.1 Limitations ... 40 7.2 Further research ... 40 8 References ... 41 9 Appendices ... 46

Appendix A. Interview guide ... 46

5

P

6

1 I

Consumer e-commerce has been growing rapidly in the past years (Weltevreden, 2008; Walker, 2015; Żurek, 2015). However, the growth of e-commerce impacts society and the environment negatively. It results in an increase in logistics traffic within cities, leading to more noise, congestion and emission (McKinnon and Ge, 2009), which deteriorates the quality of life in the city (Van Duin, 1997; Allen et al., 2012). The increase of traffic in cities is worsened by low utilisation of vehicles, leading to inefficient routings where the number of driven kilometres is relatively high.

There are several reasons for inefficient routings. At present, key factors in e-commerce are narrow time windows and low lead times (Browne and Gomez, 2011; Faccio and Gamberi, 2015), these lead to fewer scheduling opportunities and inefficient routings (McKinnon and Ge, 2009; Cherrett et al., 2017; Turkensteen and Hasle, 2017). Also, the consumer e-commerce market is fragmented, products are not consolidated, resulting in inefficient routings (Sternberg and Harispuru, 2017). Parcel carriers typically work individually, forcing the total delivery flow to be split into several smaller flows (Janjevic and Ndiaye, 2014; Sternberg and Harispuru, 2017).

In this research, we consider how negative effects on society can be reduced, since improving the sustainability of distribution within cities is a key aim nowadays (McKinnon and Ge, 2009). We aim for a reduction of logistics traffic. To achieve this, distribution should be done more efficient, which can be accomplished by higher vehicle utilisation. To be able to create higher vehicle utilisation, it is considered to combine deliveries to businesses and consumers, and integrate returns for better utilisation in the backward leg (Islam and Olsen, 2014). Returns consists of products in the city that need to be taken to a logistical facility outside the city. These include returned parcels from customer, which are increasing due to the growth of e-commerce too (Topsector Logistiek, 2017), parcels that businesses want to ship (first-mile logistics) and packaging material such as roller cages.

7

Turkensteen and Hasle, 2017; Veličković et al., 2017). It suggests to share resources of the hub among parties that have different product flows to be delivered or collected by the hub’s vehicle. Note that it is an open logistics city hub which can be used by all carrier parties. In the current logistics network, parcel carriers have several distribution centres within the country which serve as transhipment or consolidation points to create more efficient routings. Hubs have the same functionality, just outside the city centre. Are these logistical facilities also applicable within the city, to have the same benefits there? Logistical consolidation facilities in cities are known as micro-hubs. Micro-hubs enable consolidation possibilities within city centres (Mabe, 2016), and are expected to result in less driven kilometres. The applicability of micro-hubs to achieve the aforementioned goal is researched.

In this paper, it is researched how micro-hubs can facilitate in an integration of returns within the context of combined last-mile deliveries to consumers and businesses, to create more efficient routings. This reduces the amount of traffic and the corresponding negative societal impact. The goal of this paper is achieved using a design science approach. The problem is analysed and a solution is proposed. This solution is validated by means of interviews with stakeholders. It contributes to existing knowledge about sustainable and efficient logistics, which until now did not consider the use of micro-hubs within the integrated context of deliveries to consumers and businesses and returns. Also, it provides insights for stakeholders from practice.

8

2 B

First, a background on the scope of the research is given, which is an integrated form of deliveries to consumers (B2C) and businesses (B2B) using the existing network of hubs for city distribution. B2B city distribution is described in Section 2.1, B2C parcel delivery is described in Section 2.2, and both are compared in Section 2.3. The information gathered here should be taken into account when designing the solution.

Existing literature and knowledge, which is necessary to design a solution, is discussed afterwards. Returns and possibilities for integration with deliveries are discussed in Section 2.4. Information on micro-hubs is provided in Section 2.5. Lastly, in Section 2.6, involved stakeholders and their goals are described.

2.1 B2B

B2B city distribution is a concept initiated by the government when the amount of traffic to businesses in the last mile grew and became a source of nuisance within the city (Van Duin, 1997). The reason for the increase in traffic is twofold. First, businesses often have multiple suppliers, and therefore multiple vehicles have the same final delivery address. Second, deliveries often consist of relatively few goods per address, leading to low utilisation of the vehicle (Van Duin, 1997).

According to Browne et al. (2012), a reduction of traffic and distance travelled reduces negative societal impacts. This is done by the use of a hub at the edge of the city, well accessible by infrastructure and ideally close to a commercial district. In these hubs, freight is consolidated and decoupled. The consolidation enables to choose an appropriate size and type of vehicle for this load (Clausen et al., 2016), which results in higher vehicle utilisation (Triantafyllou et al., 2014). Extra handling is needed to consolidate the goods. The consolidation concept is only useful for less-than-truckload freight (Hendriks, 2017) and until now only applied in B2B, not in B2C.

9

stock keeping. Consolidation is very powerful for receivers of small loads and multiple deliveries, since they will receive everything in one delivery (McKinnon and Ge, 2009).

2.2 B2C

Since the topic of this research includes B2C parcel delivery, it is important to know more about the characteristics and challenges in this sector. Three difficulties are found: the not-at-home problem, narrow time windows and the wide dispersion of consumers.

One of the problems within B2C delivery is the not-at-home problem (Xiao et al., 2017). This problem is partly covered by contacting the consignee before delivery (Van Duin et al., 2016) and using other locations for delivery e.g. post offices, service points or local shops, where consumers can come to get their parcel themselves (Gevaers et al., 2014). However, these solutions are not always applicable as in some cases a signature is needed at the time of delivery (Gevaers et al., 2014).

Another difficulty are narrow time windows. The implementation of time windows has a negative effect on the efficiency of routings but increases the delivery success rate (Gevaers et al., 2014; Xiao et al., 2017). Nowadays, consumers are aware of the environmental issues of home delivery logistics. However, typically consumers do not want to make the trade-off between broader or no delivery windows and a more economically efficient routing.

Another problem is the wide dispersion of consumers, which often is not dense enough to make the routing economically efficient, as well as the need of enough products to be delivered (Gevaers et al., 2014; Xiao et al., 2017). Due to these problems, there is a high degree of vehicles driving with relatively low utilisation.

2.3 B2B

B2C

10

et al., 2017). To cope with these uncertainties, it is important to be able to react quickly, which makes the delivery efficiency in B2C very challenging. The chance of a high vehicle utilisation is higher in B2B compared to B2C and therefore the routing can be made more efficient (Van Duin et al., 2016).

Another difference has to do with relationships and availability. As mentioned by Van Duin et al. (2016), in the B2B market both buyers and sellers are business-oriented. Typically, their relationships and working hours are more reliable compared to customers, and often these are long-term relationships. This results in less not-at-home problems in B2B compared to B2C. Moreover, the possibility for customers to choose a timeslot in which they want their parcels delivered creates an extra challenge. This leads to less efficient routings (Van Duin et al., 2016). The same goes for same-day delivery, which leads to decreased efficiency and emptier trucks too (Faccio and Gamberi, 2015).

2.4 R

Returns can have an important role in increasing routing efficiency in the return leg. Therefore, this section provides insights on characteristics of returns, the advantages and disadvantages of integrating returns and deliveries and current procedures for return collection are described. Characteristics of returns

The growth of e-commerce created a challenge in allowing and managing product returns (Brusch and Stüber, 2013). The return flow in e-commerce is big due to the fact that consumers only have the possibility to try the product from the moment that it is delivered (Brusch and Stüber, 2013). Since the collection of returns is less time-critical than delivery, it enables to create more efficient routings. The added value of returns is typically lower, the items have the same destination when they come from the same shipper, the delivery time is more flexible and the quantity of returned products is uncertain (Chen et al., 2017).

Integrating deliveries and returns

11

However, there are also disadvantages of integrating delivers and returns. First, it disallows the two flows to function independently, and creates extra difficulties when making schedules acceptable for both parties (Tibben‐Lembke and Rogers, 2002). Second, the integration comes with physical difficulties. Loading and unloading might become more inefficient (Tibben‐ Lembke and Rogers, 2002). The most important criteria that need to be matched are location, vehicle compatibility, vehicle capacity and the time schedule (McKinnon and Ge, 2009). Many vehicle routing problem algorithms are made to study the best way to schedule these routings. The research by Turkensteen and Hasle (2017) showed that when collections and deliveries are done interchangeably, the most distance reduction and vehicle utilisation increase is achieved. The benefit is higher when the vehicle capacity is relatively small and when locations need both delivery and collection. The longer the journey is, the higher the incentive to find a suitable return load, because this can save relatively more money compared to short journeys (Verhoeven, 2008).

Collection of returns

In the Netherlands, customers mainly return their online products using service points in the city centre, the neighbourhood or a convenience centre (Weltevreden, 2008). The advantage of service points are the flexibility in delivery and collection time, its accessibility and the control of goods (Chen et al., 2017). Returns can also be delivered to unmanned locations, such as lockers (Weltevreden, 2008). Manufacturers can also directly collect the goods from customers, or they can do this via retailers or third party logistics (Agrawal et al., 2015). Third parties can work more effective and efficient as economies of scale is enabled, also individual firms do not need to make large investments anymore (Hung Lau and Wang, 2009).

2.5 M

-

To enable the combination of B2B and B2C deliveries and returns, a central logistical facility is needed, where these flows are combined. A micro-hub might be applicable for this. It is a consolidation point which usually serves smaller areas compared to normal hubs (Transport for London, 2016). It is located in the inner city and often facilitated by the city or municipality (Browne et al., 2012). It can serve as a transhipment point, from where the goods in large vehicles are decoupled into smaller loads for last-mile delivery (Clausen et al., 2016).

12

2013; Transport for London, 2016). It reduces the total distance travelled and final delivery can be done using smaller and cleaner vehicles (Browne et al., 2012; Janjevic et al., 2013). Another advantage of the micro-hub is that it enables the larger vehicles to drop off deliveries within the city centre (at the micro-hub) during night-time, while it is less busy. Currently, this needs to be done during peak hours in which the end receivers are available but heavy trucks are often not allowed to enter the city centre. Micro-consolidation typically works best for numerous small sized deliveries (Janjevic and Ndiaye, 2014) and when the distance between the current depot and delivery address is relatively high (Leonardi et al., 2012).

A disadvantage of a micro-hub is the additional handling that is created (Morganti and Gonzalez-Feliu, 2015). Due to the fact that a logistical facility is needed within the city, it is less suitable for heavy loads, which need more storage area (Janjevic et al., 2013).

The micro-hub can be designed in three ways: - A container in a parking garage,

- A trailer temporarily parked on the street (mobile hub), - A building.

The container probably leads to least hinder for the public. Parking garages are very well accessible by vans (Verlinde et al., 2012) and often located close to the city centre. The possibility to use a trailer as mobile hub differs from the container in the sense that this trailer is used to transport from the hub to the central location every morning and evening (Verlinde et al., 2014). Afterwards, the trailer can be used as a warehouse, with a loading dock for light vehicles and an office. The place to park the mobile hub should be big enough for the vehicle to manoeuvre and allow the lighter vehicle to park nearby (Verlinde and Macharis, 2016). The benefit of the building are well-accessibility for inhabitants and it decreases vacancy in the city. However, this type might not always be good accessible by vans for a longer period during (un)loading.

2.6 S

13 - Carrier: provides a transportation service,

- Government: a local authority that reflects the interests of the city, the environment and the society,

- Hub: runs the hub and does the consolidation,

14 Table 2.1 Stakeholder goals

Car rie r Gover n m en t Hu b Re ce iv er S h ip p er Reference Efficiency

Minimise travelled distance x x Gevaers et al., 2014; Faccio and Gamberi, 2015

Distance hub from the city x Van Rooijen and Quak, 2010

Critical mass needed to be financial stable and make an impact in the city

x Gevaers et al., 2014

Minimise cost x x x Harrington et al., 2016; Behrends, 2016; Faccio and Gamberi, 2015

Minimise delivery frequency per customer x x Gevaers et al., 2014

Offer low lead time x x Behrends, 2016; Harrington et al., 2016; Gevaers et al., 2014

Maximise capacity utilisation x x Gevaers et al., 2014; Harrington et al., 2016; Behrends, 2016; Cherrett et al., 2017

Consumer density level x Harrington et al., 2016; Mckinnon and Ge, 2009; Behrends, 2016

Environmental/society

Adequacy of local policy on city logistics, Sustainable Urban Logistics Policy (SULP)

x Hendriks, 2017; Van Rooijen and Quak, 2010

Strong environmental performance (green image)

15

Ambition to be energy neutral x x Mckinnon and Ge, 2009; Faccio and Gamberi, 2015; Topsector

Logistiek, 2017; Ballantyne et al., 2013

High accessibility of the city x Topsector Logistiek, 2017

Reduction of air pollution x Topsector Logistiek, 2017

Economic vitality of the city x Mangiaracina et al., 2015; McKinnon and Ge, 2009; Faccio and

Gamberi, 2015; Topsector Logistiek, 2017; Ballantyne et al., 2013; Allen et al., 2012

Improved liveability: attractive urban area, less traffic congestion and noise

x Faccio and Gamberi, 2015; Topsector Logistiek, 2017; Ballantyne et al., 2013

Flexibility

Uniform labelling x Harrington et al., 2016; Behrends, 2016; Gevaers et al., 2014

Flexibility in terms of delivery times, options and locations

x x x Harrington et al., 2016

Possibility of returning goods x x Zolfagharinia and Haughton, 2017

Cut-off time implemented by the hub x x Gevaers et al., 2014; Hopkins and McCarthy, 2016

Time windows for final delivery x x Harrington et al., 2016; Gevaers et al., 2014; Topsector Logistiek, 2017

City time slots (period in which vans are not allowed to enter the city)

x x Quak, 2008

Flexibility in planning delivery routing x Topsector Logistiek, 2017; Rogerson, 2017; Gevaers et al., 2014

16 Service level

High timeliness x x x x Hendriks, 2017

High first-time delivery success rate x x x Hendriks, 2017; Topsector Logistiek, 2017

Parcel traceability x x x Gevaers et al., 2014; Allen et al., 2017

Receiving proof of delivery x x x Harrington et al., 2016; Ballantyne et al., 2013; Allen et al., 2017 Transparency in costs, benefits and

reasoning

x Gevaers et al., 2014; Iwan et al., 2016; Harrington et al., 2016; Hendriks, 2017

Customised delivery service x x x Topsector Logistiek, 2017; Harrington et al., 2016; Allen et al., 2017 Minimise effort to collect parcels, home

delivery for business receivers

x Harrington et al., 2016

17

3 M

3.1 R

This research is done following the design science research paradigm (DSR). DSR is a research method that produces an artefact which is created to address a problem (Hevner et al., 2004; Peffers et al., 2007). This should be an important business problem which has not been solved yet (Peffers et al., 2007). In this research, the aim is to find a solution to the practical problem that is introduced in Chapter 1: the increase of logistics traffic in cities due to the growth of the e-commerce sector. Developing the solution to the problem should be a process that uses existing theories and knowledge to create a solution to the problem defined (Peffers et al., 2007).

Peffers et al. (2007) created a process model that indicates six activities that need to be done in DSR, see Figure 3.1. Also, it shows several entry points for the research. In this case the research starts with the first activity, since it is a problem-centred initiation. It is important to notice that the activities are done in sequence, but it also is an iterative process. After evaluation and communication it might be needed to go back to defining objectives or designing.

Figure 3.1 DSR Process model (Peffers et al., 2007)

18 Figure 3.2 Regulative Cycle (Wieringa, 2009)

The regulative cycle does not include a communication activity like the DSR process model does. It seems that Peffers et al. (2007) do not include solution validation which contradicts the regulative cycle. The DSR process model continues with demonstration of the solution after designing. Afterwards, the solution is evaluated. Solution evaluation is a critical part of DSR, which provides evidence for the solution to work and achieve the purpose that is was designed for (Venable and Pries-heje, 2012). Since the DSR process model does not validate the solution before implementing, the regulative cycle seems to be more beneficial and is therefore used in this research.

The exploitation of steps 1, 2 and 3 are discussed in the next sections. Step 4 and 5 are too expensive and time-consuming to exploit within the scope of this research, which is common for students (Wieringa, 2007). To cope with these limitations, validation includes as many influences from practice as possible, by doing interviews with stakeholders from practice. We advise to do implementation by a pilot project within one city, to test the solution under real conditions of practice (Wieringa, 2009). Since it is not sure how successful the solution will be, this is preferred over a big-bang strategy. In case it is successful, the pilot should be enlarged to more cities.

3.2 P

Problem investigation asks for information and understanding of the experienced problem (Wieringa, 2009). The scope and the problem have already been discussed in Chapter 1. Information about the problem is gathered from literature and validated during meetings with Hendriks, an expert in city distribution. Also stakeholder goals within this scope as mentioned in Section 2.6 are validated this way.

3.3 S

19

the solution can meet the requirements of the stakeholders. Discussions with Hendriks are done to shape the ideas for the solution design. The solution is presented in Chapter 4.

3.4 D

Validation in DSR has multiple purposes, of which a key is to tell how the solution scores (Wieringa, 2007; Venable and Pries-heje, 2012) and how it helps stakeholders to achieve their goals (Wieringa, 2009). Validation is a crucial component of the research process which can provide essential feedback to the design (Hevner et al., 2004), as one of the goals is to determine areas for improvement (Venable and Pries-heje, 2012). Therefore, design validation is emphasised in this research. The solution should satisfy the requirements and constraints of the practical problem to be effective (Hevner et al., 2004). Important questions for validation are (Wieringa, 2009):

- Internal validity: does the solution satisfy the criteria?

- External validity: would this solution satisfy the criteria when implemented in a different context?

Validation method

To get the best validation results, the solution is explored within its real environment, a natural setting. We conduct ex post evaluation, as we evaluate after constructing the solution (Venable and Pries-heje, 2012). According to the DSR Evaluation Method Selection Framework by Venable and Pries-heje (2012), suitable methods for natural ex post evaluation are action research, case study, focus group, participant observation, ethnography, phenomenology and survey. Since the aim is to gain in-depth knowledge on how stakeholders assess the solution, case study and focus groups are the best options. Case study interviews would suit best, because in focus groups interviewees can influence each other, leading to biased results (Hevner, 2014). A cost-benefit analysis is also a type of validation method that could be useful for this research (Hevner et al., 2004). However, as the focus is on the effects for the practical problem, which is not cost-based, this is out of scope but it might be interesting for the feasibility of the solution. Case selection

20

of biased interpretations and creates a higher chance of generalisation of findings (Karlsson, 2009). An extra interview is done with an expert on returns, to get more information on this part of the solution. For the list of interviewees, see Table 3.1.

Table 3.1 Interviewees

Date Stakeholder type Organisation Interviewee(s) 8-12-2017 Carrier Go-Fast Bicycle

Delivery Services

Peter Rugge

18-12-2017 Carrier PostNL Maryam Boonstra, Marien Vaandrager

5-12-2017 Government Municipality of Groningen

Sjouke van der Vlught, Jeroen Berends, Mario Sabel

7-12-2017 Government Province of Drenthe Rolf Meerbach 18-12-2017 Hub Stadsleveransen Göteborg Christoffer Widegren 20-12-2017 Hub Binnenstadservice Nijmegen Frank Adema 20-12-2017 Hub Binnenstadservice Nijmegen Birgit Hendriks 7-12-2017 Hub Binnenstadservice Maastricht Max Prudon 14-12-2017 Hub Noorderpoort Roeland Hogt

20-12-2017 Receiver B2B Ideska Arnoud Schoffelmeer 22-12-2017 Receiver B2C - Marijn te Flierhaar 14-12-2017 Receiver B2C - Hugo Herbers 22-12-2017 Shipper The Musthaves Steven de Boer 15-12-2017 Shipper Bol.com Hessel de Gelder 15-12-2017 Shipper Blokker Roel Megens 18-12-2017 Expert returns BuyBay Teun Kraaij

3.5 D

21

collection method is used, which means that there is no triangulation (Karlsson, 2009). This reduces the reliability of the data as the same weaknesses might be applicable to all data. To overcome these possible weaknesses and guarantee unbiased data interpretation, the interviews are held in pairs of two researchers so that different interpretations can be discussed (Karlsson, 2009). Also, the interviews are recorded and transcribed to be able to work with primary data. None of the interviewees felt the need to verify the interviews. In total, sixteen interviews are done with a duration of approximately one hour.

The interviews start with an introduction. In the body of the interview, several topics on the content are discussed. Questions are asked about the general opinion on the solution, the expected advantages and disadvantages, the effect on the interests of the stakeholder, the feasibility of the solution and how this would contribute to solving the problem. To conclude, the interviewee is asked for extra topics and subsequent steps are explained. The interviews are semi-structured, as that is the best method to find answers on specific questions while being open for extra information that is proposed by the interviewee (Myers, 2010). The questions are mainly open formulated, to enable the interviewee to give as much information as possible. Most interviews are face-to-face, within the working environment of the interviewee. However, due to travelling distance and availability in time some of the interviews are done using Skype or FaceTime. This has not resulted in any restrictions.

It is important for internal validity to discuss each variable that has an influence on the assessment. To guarantee construct validity, all concepts should be well defined to guard from misunderstandings (Karlsson, 2009). As only stakeholders within this industry are considered, it is difficult to check the validity within another environment.

3.6 D

The gathered raw data is qualitative. The data is coded using the software ATLAS.ti to analyse and structure the data. We want to investigate what new information emerges from the interviews and what is in line with findings from literature.

4 S

The proposed solution is designed to solve the problem of the amount of traffic in cities, which is increasing due to the growth of e-commerce. The goal is to increase vehicle utilisation for more efficient routings. The scope in which the solution is proposed, is an integrated distribution network of deliveries (B2B and B2C) and returns. The goal should be achieved under the conditions of societal focus and a high customer service level (lead time, flexibility, easiness of returns). As the focus is to create a more sustainable network, most focus is on sustainability and less focus is on costs. However, all categories of stakeholder requirements in Table 2.1 are taken into account to design the solution.

We propose the following solution: using a micro-hub as logistical facility in the city (either a building, container or mobile hub), which is open to all logistics parties, referred to as white-label. The micro-hub servers as starting point from where deliveries can be made to consumers and businesses, and returns can be collected. This is done using bicycle couriers. Also, the micro-hub serves as a service point where consumers can collect or return parcels themselves. Therefore, it serves as both a vehicle reception point and a goods reception point (Janjevic and Ndiaye, 2014). Depending on the size of the city, multiple micro-hubs are used.

In this chapter, more aspects of the micro-hub are discussed (Section 4.1). Next, it is explained how the delivery (Section 4.2) and return collection (Section 4.3) are done. Finally, white-label is explained and discussed in Section 4.4.

4.1 M

-

The micro-hub serves as multifunctional logistical centre in this solution. The functionalities of the micro-hub are as follows:

- Sorting area,

- Transhipment point from van to cargo-bikes, - Temporary storage of deliveries,

23 Micro-hub types

The micro-hub can be designed in three ways, as explained in Section 2.5: a container in a parking garage, a trailer temporarily parked on the street (mobile hub) or a building in the city. Until now, little information is known about the advantages and disadvantages per type. There is not yet a strong preference. Therefore, all three types are considered in the solution and questioned during the validation interviews, from which insights can be gained.

Micro-hub supply

One of the main reasons to choose a micro-hub is the fact that the time that vans are in the city decreases, since there is no need to stop at multiple addresses and roll containers containing multiple parcels can be unloaded very quickly at the micro-hub, making the urban environment more liveable and sustainable (Janjevic et al., 2013; Mabe, 2016). This also reduces the tension created by the city time slots. Deliveries from hub to the micro-hub can be done in evenings or during the night, leading to less congestion and disturbance (Browne et al., 2011).

Bicycle delivery

The last mile is done by bicycle couriers, which are environmentally friendly and less hindering due to their speed and space requirements (Browne et al., 2011; Janjevic et al., 2013). They are suitable for busy areas (Mabe, 2016) and significantly cheaper compared to cars (Joerss et al., 2016). Due to the small volumes and weights, high consumer density and low distance from micro-hub to final address, the applicability is high. Flexibility is created by the short distances to travel and the possibility to enter city time slot areas with cargo-bikes. Furthermore, the relative low capacity of cargo-bikes is no problem, as it is possible to return to the micro-hub that is always nearby. We see that this results in an increase of driven kilometres in cities, but as they are done with emission free vehicles, this should not be a problem (Browne and Gomez, 2011). Within urban areas, the speed disadvantage compared to cars is negligible (Joerss et al., 2016).

Limitations

24

attractiveness of the micro-hub therefore is a critical factor (Janjevic and Ndiaye, 2017). Another challenge is the departure time at the hub, which should not be too late and not too early. To be most efficient, all parcels are transported to the micro-hub in the same vehicle. A trade-off should be made in waiting for the last parcel to arrive or starting delivery earlier.

4.2 D

A micro-hub suits the concept of combining consumers and businesses parcels, due to the fact that a critical mass of volume is needed for the micro-hub to work effectively (Gevaers et al., 2014). Furthermore, it has a high effect on vehicle reduction which is beneficial for society, as without consolidation a lot of vehicles would be needed to deliver all parcels from different shippers and carriers to the customers (Sternberg and Harispuru, 2017).

A difficulty in combining consumer and business deliveries is their non-matching availability (Van Duin et al., 2016), making it almost impossible to deliver both in the same routing (Hendriks, 2017). Therefore, the micro-hub serves as a decoupling point, where consumer and business deliveries can be divided in different delivery routings, but still benefit from the combined transportation to the inner city.

Delivery flow

25 Figure 4.1 Delivery flow

4.3 R

In the solution, return and deliveries are done at the same location during the same routings, as this leads to best utilisation of vehicles (Turkensteen and Hasle, 2017). This is possible as businesses and consumers want to both receive and return parcels. The final destination for returns is the same (Chen et al., 2017), within the scope of this solution the hub. A pre-condition is the design of vehicles, which should counteract the difficulties that combined loading and unloading may bring (Tibben‐Lembke and Rogers, 2002). By the use of digital platforms, consumers and businesses can indicate when preferably returns need to be collected. This information is used to create schedules for the bicycle couriers and the van driving from and to the hub.

Another benefit of the micro-hub is the easiness of the return logistics that is enabled for self-delivery of returns. The advantage of a micro-hub over service points in e.g. supermarkets and book shops is the fact that the micro-hub is specialised in these processes. Therefore, collecting or returning a parcel is done quickly (Kauf, 2016). Furthermore, there is enough space to store parcels, which do not hinder anyone from daily business.

Return flow

26

stored at the micro-hub after it unloads at the micro-hub (3b). The van is also collecting high volume and weight returns at companies and consumers (3a, 3c, 3d) and returns to the hub with returns to be transported elsewhere in the country (2). These can be collected at the hub, by corresponding carriers that are supplying goods to the hub (1).

Figure 4.2 Return collection flow

4.4 W

-

The micro-hub is open for all logistics parties, referred to as white-label. This results in benefits for several stakeholders.

The micro-hub can benefit from economies of scale (Hung Lau and Wang, 2009) as the network is bigger, it is easier to find volume that can be used to maximise utilisation of the vehicle and make efficient routings. Furthermore, small retailers are enabled to join a fast delivery and returns network without making huge investments (Mabe, 2016). Also, the most expensive part of logistics can be outsourced for carriers, which is low volume last-mile delivery (Mabe, 2016). Finally, it results in more convenience for the receiver, as delivery is done only once a day, not by multiple carriers (Kauf, 2016).

27

5 D

In this chapter, the findings of the interviews are discussed. These consider the stakeholders’ opinions on the solution, its advantages, disadvantages and feasibility. The interview data is coded into categories and five corresponding themes, see Table 5.1. A more extensive version of the coding tree can be found in Table 9.2 in Appendix B. The findings within each theme are discussed in the following sections. These are used to consider whether the solution meets the requirements to achieve the goal.

Table 5.1 Coding structure

Third order themes Second order category

Efficiency oriented Costs: delivery, consolidation, micro-hub, returns

Efficiency: dispersion of final addresses, success rate, additional handling, vehicle utilisation

Society oriented Number of vehicles: consolidation, micro-hub, returns City liveability: consolidation, micro-hub

Emission and sustainability: consolidation, delivery, micro-hub Enables bicycle delivery

Customer oriented Customer effort: delivery, returns Flexibility: delivery, returns

Service level: delivery, returns, lead time, success rate Recognisability

Time windows

Willingness to pay: delivery, returns Design micro-hub Extra services offered

Availability

Combination with other solution Design: building

Design: mobile hub Design: container Location micro-hub Location returns Recognisability

28 Feasibility of the

solution

Business model Reasons to join

Collaboration: agreements with carrier, carrier collaboration, competition web shops, differentiation strategy, relationship carrier/retailer -

customer, stimulation collaboration

Compatibility to combine products: delivery, returns, loading/unloading, schedules, same destination returns, vehicle capacity

Liability: delivery, returns Accessibility for vans

Applicability bicycle delivery Government actions/restrictions

Returns: added value, time-consuming, time-critical

5.1 E

Findings within this category consider routing efficiency and costs.

Routing efficiency. In general, the interviewees indicate that consolidation increases routing efficiency as products to last-mile areas can be combined. Also, it is more efficient that the van only needs to supply the micro-hub. “Personally I like this idea, as you deliver at only one location with large traffic.” (Municipality of Groningen).

It is mentioned by all stakeholder groups, except the government who did not discuss this topic, that providing a self-collection option probably results in more flexibility for the receiver, less failed deliveries and thus higher routing efficiency. However, extra handling is created at the micro-hub, which is not efficient. “I don’t know whether it is optimal to repack the parcels.” (BuyBay). But the extra handling is not applicable to the mobile hub. “You don’t need to move it from the truck to a hub, you will do it directly from the truck.” (Bol.com).

Costs. Consolidation reduces costs. “It is consolidated, you can have six stops in one, so costs

29

tremendously cost increasing.” (Binnenstadservice Hendriks). “That will not cost a lot extra.” (Binnenstadservice Adema). It is discussed that self-collection is not profitable, unless clients buy something in your shop, which is not offered in the micro-hub.

Conclusion. The solution creates higher routing efficiency, which reduces costs. However, building a micro-hub is expensive, as well as the labour costs for availability.

5.2 S

Findings within this category consider effects on traffic and the applicability of bicycle delivery.

Traffic. It is mentioned by all parties that the solution is beneficial for society, as the amount of traffic reduces by consolidation and the use of bicycle couriers. “Without a doubt you will remove traffic movements from the city.” (Binnenstadservice Adema). By this traffic reduction, city liveability increases. “For the liveability I think it is good to reduce that [traffic] as much as possible.” (Province of Drenthe). However, extra consumer traffic is created due to self-collection. Furthermore, some state that including returns in consolidation at the micro-hub has little impact on the number of vehicles, “They [waste collection trucks] drive with two or three trucks, but it is only once a week. I don’t think that is the gain.” (Binnenstadservice Adema).

Bicycle delivery. Another societal benefit is that bicycle couriers create les hinder compared to vans in terms of parking space and congestion. “I think the receiver gets a positive feeling, as it is greener and disturbs less at pavements.”(Bol.com). However, it it reasoned that cities are not yet prepared for an increase of bicycle couriers which can create unsafe situations. “The number of cargo-bikes in the city will grow and many cycling paths are not designed for big cargo-bikes.” (PostNL).

Conclusion. The feasibility of the solution for the society is sufficient as benefits are achieved. However, the effect of including returns on traffic should be further analysed. Also, something should be done to enable more cargo-bikes in the city centre.

5.3 C

Findings within this category consider self-collection of parcels, customer effort and service level.

30

can collect and return parcels, it gives customers control over the process.” (Binnenstadservice Prudon). However, the self-collection possibility is questioned too, as it is expected by all stakeholder groups that consumers do not like to make effort for parcel collection. “A great convenience of online shopping is home delivery.” (Receiver Te Flierhaar). All stakeholder groups think that it would only work under certain circumstances, such as low distance, broad opening hours, the need to have the parcel quickly, being aware of the negative effects of home delivery or when this is cheaper or free of charge. “When I order a parcel, I want to have it quickly. It does not necessarily need to be delivered at my home, I like to be able to collect it somewhere.” (Province of Drenthe). Businesses might be willing to collect parcels before opening hours, to have them available in their shop earlier. Having a return collected at home is seen as effort by some stakeholders too, as one needs to stay home for it, while others do not see this as an effort. “It results in more convenience for the customer, he does not need to leave the house for a return.” (PostNL). “I think that collecting returns is difficult, because people need to stay home for something they don’t want anymore.” (BuyBay).

Service level. Customers prefer face-to-face contact and flexibility in moments to receive parcels. “If that point closes at 9, you need to get there before 9.” (Receiver Herbers). Also, they prefer to have one consistent location for both deliveries and returns. “You should not need to visit Jumbo one time and another small shop the other time.” (Receiver Te Flierhaar). A disadvantage of the micro-hub in terms of service level is the lead time, which can be delayed by one day if a parcel is delivered late at the hub. “Next day delivery is possible if the courier delivers on time at the hub.” (Binnenstadservice Adema). On the other hand, cargo-cycles can move faster within urban areas.

Conclusion. To have a feasible solution from the customer perspective, it is important to offer flexibility and face-to-face services. Therefore, it would be good to have a self-collection point at the micro-hub. However, this contradicts with the effort that customers want to make. Mostly customers still prefer high service home delivery.

5.4 D

-

Findings within this category consider the micro-hub type, its availability and location.

31

need, while others say that there should be a canteen for carriers. “It is not more than an expedition room where products are delivered.” (PostNL). Mobile hubs and containers are perceived to be ugly, but others state this is no problem in parking garages. “Parking a car somewhere and deliver from there, that looks like shit.” (Binnenstadservice Prudon). Advantages of a building, mentioned by the interviewees, are promotion possibilities, guaranteed availability, decrease of vacant properties and convenience. A big disadvantage are the high costs for the building.

Availability. Some interviewees say that all day availability of the micro-hub is unnecessary, while others state there is an absolute need to have broad opening hours. “When you do that all day, you need to be manned all day, which is nonsense.” (Binnenstadservice Adema). It is proposed to exploit the micro-hub within existing public areas (e.g. library) instead of having an own facility, to easily create availability.

Location. The micro-hub should be located within an accessible area, where people are anyway, but outside the city time slot area. “If a truck still need to go into the city time slot area, you are missing the point of zero emission.” (Go-Fast). Consumers prefer one place for collection and returns, but it is possible to return at offline shops too. “As we have 500 shops in the Netherlands, you can always return it to the shop, after which I take it back through my own network.” (Blokker). Generic labelling systems should not be a problem as they are available. However, not all stakeholders are aware of the existence. “The tracking of parcels continues in our process, so I guess that is no complication.” (Binnenstadservice Prudon).

Conclusion. No big feasibility problems have arisen. However, it should be considered what opening hours are needed and how the advantages and disadvantages of the design types can be rated.

5.5 F

Findings within this category consider the business model, collaboration between stakeholders, combining delivery and returns and bicycle delivery.

Business model. Many stakeholders question the business model, who will pay for this? “The

32

to create a business model.” (Province of Drenthe). Interviewees indicate that the project should also not depend on subsidies. “I guess you make a dangerous business model when you depend on subsidy instead of your customers.” (The Musthaves). No attention is yet given to this topic as it is not the scope of this research.

Collaboration. Collaboration seems to be a big barrier for this solution. Some carriers do not want to share service points based on differentiation strategies, while others expect this to change in the future. “That is very unique per company, and that is how they want to differentiate.” (BuyBay). One outstanding reason not to collaborate, is that multiple competing companies keep each other sharp and therefore increase efficiency. “I think if you let it to two commercial parties […], if one does better than the other, the other wants to beat it, in the end it gets more efficient.” (Go-Fast). Hub parties indicate that big shippers can motivate carriers to do something different. A hub cannot initiate change, but businesses and shippers can. “I am sure that bigger web shops talk to carriers convincing that something needs to change or it will get stuck.” (Binnenstadservice Prudon). No attention is yet given to this topic as it is not the scope of this research.

Integration deliveries and returns. Combining delivery and returns is tricky in terms of timing, since returns are not always ready at the moment of delivery. “We deliver one time at the day, and the outbound goods are ready for pickup later at the day.” (Stadsleveransen Göteborg). To solve this, the consumer/business can be asked to prepare the return, so that it can be collected at the same moment, which is also beneficial for the consumer/business. “The more you can do in the same moment, the less time it takes.” (Ideska). Furthermore, there are some small problems on the practical side. The return needs to be packaged in a way that it can be transported safely. “Is it packaged in a way that it can be transported? That can be a disadvantage.”(PostNL). Also, it is not convenient to handle both parcels and pallets within one vehicle. “These flows [parcels and pallets] cannot be combined.”(Binnenstadservice Prudon). Therefore parcels should be transported on a pallet in the van and only parcels are delivered by bicycle.

33

to cover the volume of a van. Another disadvantage of a cargo-bike is the lower security compared to vans. “If I need to go into the street with cigarettes, which represents value, you can do it best by car.” (Binnenstadservice Adema).

Conclusion. The solution is feasible under the condition that a healthy business model is set up and carriers are willing to collaborate and share service points. This can be achieved by the influence of big shippers. To combine delivery and returns, there are some small practicalities that need to be considered. Bicycle delivery is feasible, but many cargo-bikes and couriers are needed.

5.6 V

Whether the solution satisfies the criteria is the most important question for internal validity (Wieringa, 2009) and is answered below. The aim of the solution is to increase vehicle utilisation and therewith reduce the number of driven kilometres in cities, by more efficient routings. This should be achieved under the conditions of societal focus and a high customer service level.

It is agreed by all stakeholder type that the goal is achieved, as the number of vehicles are reduced by this concept and therewith also the societal problems (emission, congestion etc.). However, an increase of vehicles is expected due to consumer trips for self-collection. Furthermore, the effect of including return collection on traffic is expected to be low, which is a focus point in the solution.

The condition of high customer service level is achieved. Lead time might increase due to waiting time for consolidation. However, it can be decreased due to the use cargo-bikes instead of vans. Flexibility is considered by offering a self-collection option. Furthermore, the easiness of returns is accomplished by home pickup.

External validity

34

From the interview with Blokker, it emerged that omni-channel companies use their offline shops as a point for full truckload delivery and self-collection. “Offline shops are perfect hubs, because full truckload is delivered there.” (Blokker). “There are over 500 Blokker shops in the Netherlands, so I have delivery points… hubs!” (Blokker). Therefore, the concept of self-collection at the micro-hub might not be attractive to omni-channel companies. However, consolidation at a micro-hub can still be beneficial for these companies in terms of efficient and sustainable last-mile delivery.

35

6 D

This chapter reflects on the findings, how these relate to literature and suggestions for redesign are given. One of the main points that is discussed is the effect on traffic, as this is the main goal of the solution. Furthermore, the self-collection point and the business model are discussed, as these created discussion and are seen as barriers or improvement possibilities. Other findings are shortly related to literature.

6.1 E

Many interviewees confirmed that the amount of traffic decreases by employing this concept. It is already known from literature that consolidation in micro-hubs reduces traffic (Conway et al., 2011). However, in this research returns were included too, and the effect of that on the amount of traffic is questioned in the interviews, as returns are currently already consolidated at service points. But from vehicle routing problems we have learned that combining deliveries and returns in one routing reduces the number of kilometres in total (Turkensteen and Hasle, 2017). Literature and interviews are contradicting on this topic and we expect, based on literature, that collection of returns does reduce traffic to some extent.

Furthermore, the self-collection option leads to an increase of consumer trips to the micro-hub for parcel collection. Upfront, this was not considered as a disadvantage, but it can be confirmed by literature (Morganti et al., 2014; Van Duin et al., 2016). This negative side-effect is higher in rural areas compared to urban areas (Morganti et al., 2014). We therefore conclude, that self-collection at a micro-hub is better applicable in urban areas, where consumers need to travel only a shorter distance, resulting in less car trips (McLeod et al., 2006).

6.2 S

-

36

Furthermore, the profitability of a self-collection point is questioned, as the costs for the facility and labour are expected to be high. During the interviews it is advised to offer extra services or products, from which money can be earned. This is comparable to what is currently done in service points, where money is mainly made based on the extra traffic in the shop and a small fee is received for each handled parcel (Weltevreden, 2008). As far as known, there are no service centres that offer parcel collection and return options, without offering extra services. In this context, services should be offered to both businesses and consumers.

6.3 B

In general, it is expected by the interviewees that the business model of this solution will be difficult. This confirms what is stated in literature before about consolidation initiatives, the costs of the additional transhipments are not covered by financial gains of consolidation (Janjevic et al., 2013). Money is mainly needed for location and labour (Conway et al., 2011), these factors were considered as expensive in the interviews too, in choosing the type of micro-hub design and manning the micro-micro-hub for availability.

Several interviewees indicated that the business model should not depend on subsidies, which is confirmed by Allen et al. (2012), who state that a hub should be financially stable in the medium- to long-term. However, Janjevic and Ndiaye (2014) state that the higher the level of financial support from institutional actors, the higher the chance of success. It is unlikely that major developments can proceed without initial funding (Allen et al., 2012), only a few hubs have survived without subsidies (Janjevic et al., 2013). The government could also trigger sustainable logistics by other incentives, such as lower tax or exemptions for access to certain areas (Kauf, 2016). These incentives are expected to be able to contribute to the business model in this context too.

We can contribute that the business model is expected to be as difficult in this context compared to only business deliveries as studied until now. The only difference is that there is a possibility to add customer services to make extra money, which is currently only done for B2B services, such as stockholding and waste collection (Allen et al., 2012).

37

Interviewees indicated that a micro-hub designed as container or mobile hub would not work convenient. The only thing known in literature about convenience is that drivers prefer working from their own depot (Verlinde et al., 2014).

Combination delivery and returns

It is confirmed in the interviews that no practical problems would occur in combining deliveries and returns in one routing, under the condition that carriers work precisely, returns are packaged in a convenient way and either pallets or parcels are transported in a vehicle. The difficulties in creating schedules mentioned by Tibben‐Lembke and Rogers (2002) are not confirmed during the interviews, as long as it is known beforehand where to collect returns. Bicycle couriers

During the interviews it is discussed that a growing number of cargo-bikes is unsafe for other road users, while at first instance cargo-bikes are perceived to be safer than trucks (Conway et al., 2011; Leonardi et al., 2012). Conway et al. (2011) state that introducing unfamiliar vehicles within city infrastructure can create unsafe situations. Interviewees suggested to increase the size of the cycling path. However, Janjevic et al. (2013) indicate that the effect of a micro-hub is not sufficient to justify new infrastructure from an economic perspective. However, this might be the case as in this context micro-hubs cover B2B, B2C and returns, and thus have more effect.

Lead time

In literature it is stated that customers prefer to have a low lead time (Harrington et al., 2016). However, findings show that customers do not highly value this. Anticipating on this could result in more efficient routings.

6.5 R

In this section, it is explained what adjustments are proposed during the interviews. These can be used as after validation, the regulative cycle attempts to go back to solution design. However, these are only hypotheses and will not be tested within this research.

38

 Exploit the micro-hub in public buildings with broad opening hours, employees who are available can help customers in the self-service (Binnenstadservice Hendriks).  Make a combination with current service points or entrepreneurs, instead of

self-collection at the micro-hub (Stadsleveransen Göteborg, Binnenstadservice Max).  Add lockers to the micro-hub, to create flexibility during unmanned hours

(Municipality of Groningen, The Musthaves, Noorderpoort).

 Offer extra services at the micro-hub to increase customer value and create extra revenue. Examples are packaging of returns, temporary storage for consumers during shopping, a mini-shop, or selling coffee (Municipality of Groningen, Province of Drenthe, Bol.com, Blokker, BuyBay, Binnenstadservice Adema).

39

7 C

In this research, the applicability of a micro-hub in an integration of deliveries (B2B and B2C) and returns is considered. The aim is to improve routing efficiency for more society friendly city distribution. This aim resulted from the practical problem of an increasing amount of traffic and nuisance within cities, due to the growth of e-commerce logistics.

The research is done using the design science approach. A solution is made based on existing literature and conversations with an expert on city distribution. The solution is validated by sixteen interviews with stakeholders from different stakeholder groups: carriers, government, hub, receivers and shippers. The qualitative data is coded in three levels, categorised and analysed.

The findings show both successes and improvement possibilities.

- It is confirmed that the proposed solution is expected to improve efficiency of routings due to consolidation at the micro-hub. Therewith it reduces traffic and societal negative effects. However, the effect of including returns on traffic reduction is expected to be low during interviews, but literature states that this can be successful.

- Bicycle delivery is confirmed to be very applicable in this situation. However, cities need to prepare infrastructure for this.

- Returns and deliveries can be integrated within one routing, only small practical problems emerged which can be solved easily. The consumer/retailer needs to have the return ready on time.

- Returns can be delivered at the micro-hub by customers, as the effort that customers want to make for returns is perceived to be higher compared to deliveries. Including a return service at the micro-hub provides clarity and recognisability, which is valued by customers.

- Exploiting the micro-hub as a self-collection point for parcels is not yet very satisfying, due to the high costs and low revenue that come with it. Also, customers do not like to make effort for parcel collection. However, they do like flexibility and multiple options. - It is mentioned by both interviews and literature that high costs for the facility and availability of the micro-hub are a barrier. In addition, the business model is still difficult as no extra revenue is created.

40

7.1 L

There are several limitations to this research. Only two carriers are interviewed, PostNL and Go-Fast. These are two completely different carriers as PostNL is very big and Go-Fast is much smaller and delivers using cargo-bikes, they both have different opinions. We did not speak to a smaller carrier party that delivers by vans, which might have other opinions. Next to that, the selection of receivers is relatively small, taking into account that receivers have different requirements and preferences. Furthermore, we did not speak to service points (e.g. Bruna and Primera) which could be helpful to create a better understanding of the feasibility of self-collection at the micro-hub.

7.2 F

This research ends after validation and proposes ideas for a redesign. As design science research is an iterative process, we advise to go back to the phase of solution design and validation again, to redesign based on the proposed adjustments and validate the adjusted design. When the solution meets the validation requirements, it can be implemented.

A cost-benefit analysis is a type of validation method that would be appropriate and helpful for the feasibility of this solution (Hevner et al., 2004). Therefore, we advise to make a business case that considers all (financial) effects on the supply chain, in both short- and long-term. This can provide more clarity on the costs and benefits for each party, and might be helpful in building a healthy business model. Also, the willingness of consumers and businesses to pay for more sustainable delivery should be researched and can complement the business model. During one of the interviews it is mentioned that the concept might also be applicable to low-density areas. These areas currently cope with very inefficient routings. Having a central point for delivery might decrease driven kilometres in rural areas too. Although this is out of scope for this research, it might be interesting to do research on in the future.

The way of shopping is currently changing from offline shopping to e-commerce, but what might happen in the future? It is expected that this will change. Imagine only online shopping is done and offline stores do not have inventory, but deliver products to consumer houses. It results in a shift to only B2C logistics, as shops do not need replenishment, and likely an increase in returns. What can the role of a micro-hub be in this situation?

41

8 R

Agrawal, S., Singh, R. K. and Murtaza, Q. (2015) ‘A literature review and perspectives in reverse logistics’, Resources, Conservation and Recycling, pp. 76–92.

Allen, J. et al. (2012) ‘The role of urban consolidation centres in sustainable freight transport’, Transport Reviews, pp. 473–490.

Browne, M. et al. (2012) ‘Reducing social and environmental impacts of urban freight transport: A review of some major cities’, Procedia - Social and Behavioral Sciences, 39, pp. 19–33.

Browne, M., Allen, J. and Lenoardi, J. (2011) ‘Evaluating the use of an urban consolidation centre and electric vehicles in central London’, IATSS Research, 35(1), pp. 1–6.

Browne, M. and Gomez, M. (2011) ‘The impact on urban distribution operations of upstream supply chain constraints’, International Journal of Physical Distribution & Logistics Management, 41(9), pp. 896–912.

Brusch, M. and Stüber, E. (2013) ‘Trends in logistics in the German E-commerce and the particular relevance of managing product returns’, LogForum, 9(4), pp. 293–300.

Chen, C. et al. (2017) ‘Using taxis to collect citywide E-commerce reverse flows: a crowdsourcing solution’, International Journal of Production Research, 55(7), pp. 1833–1844. Cherrett, T. et al. (2017) ‘Logistics impacts of student online shopping - Evaluating delivery consolidation to halls of residence’, Transportation Research Part C, 78, pp. 111–128.

Clausen, U., Geiger, C. and Pöting, M. (2016) ‘Hands-on testing of last mile concepts’, in Transportation Research Procedia, pp. 1533–1542.

Conway, A., Fatisson, P. and Eickemeyer, P. (2011) ‘Urban micro-consolidation and last mile goods delivery by freight-tricycle in Manhattan: Opportunities and challenges’, TRB 2012 Annual Meeting. Washington, D.C., USA.

Van Duin, J. H. R. (1997) ‘Evaluation and evolution of the city distribution concept’, Transactions on the Built Environment, 30, pp. 327–337.

42

Estrada, M. and Roca-Riu, M. (2017) ‘Stakeholder’s profitability of carrier-led consolidation strategies in urban goods distribution’, Transportation Research Part E: Logistics and Transportation Review, 104, pp. 165–188.

Faccio, M. and Gamberi, M. (2015) ‘New city logistics paradigm: From the “Last Mile” to the “Last 50 Miles” sustainable distribution’, Sustainability (Switzerland), 7(11), pp. 14873– 14894.

Fontaine, P. et al. (2017) ‘The impact of combining inbound and outbound demand in city logistics systems’, Proceedings - International Computer Software and Applications Conference, 2.

Gevaers, R., Van de Voorde, E. and Vanelslander, T. (2014) ‘STRAIGHTSOL project: Strategies and measures for smarter urban freight solutions’, Cahiers de l’Observatoire de la mobilité, 1(1), p. 17.

Harrington, T. S. et al. (2016) ‘Identifying design criteria for urban system last-mile solutions -A multi-stakeholder perspective’, Production Planning and Control, pp. 456–476.

Hendriks, B. (2017). Interview with city distribution expert.

Hevner, A. (2014) ‘Design science research’, in Computing Handbook, Third Edition, pp. 22-1-22–23.

Hevner, A. R. et al. (2004) ‘Design science in information systems reserach’, MIS Quarterly, 28(1), pp. 75–105.

Hung Lau, K. and Wang, Y. (2009) ‘Reverse logistics in the electronic industry of China: a case study’, Supply Chain Management: An International Journal, 14(6), pp. 447–465. Islam, S. and Olsen, T. (2014) ‘Truck-sharing challenges for hinterland trucking companies: A case of the empty container truck trips problem’, Business Process Management Journal, 20(2), pp. 290–334.

Janjevic, M., Kaminsky, P. and Ndiaye, A. B. (2013) ‘Downscaling the consolidation of goods-state of the art and transferability of micro-consolidation initiatives’, European Transport, 4(54), pp. 1825–3997.

43

Janjevic, M. and Ndiaye, A. B. (2014) ‘Development and application of a transferability framework for micro-consolidation schemes in urban freight transport’, Procedia - Social and Behavioral Sciences, 125, pp. 284–296.

Joerss, M. et al. (2016) ‘Parcel delivery: The future of last mile’, Travel, Transport and Logistics, (September), pp. 1–32.

Karlsson, C. (2009) Research methods for operations management.

Kauf, S. (2016) ‘City logistics - A strategic element of sustainable urban development’, Transportation Research Procedia. Elsevier B.V., 16, pp. 158–164.

Kedia, A., Kusumastuti, D. and Nicholson, A. (2017) ‘Acceptability of collection and delivery points from consumers’ perspective: A qualitative case study of Christchurch city’, Case Studies on Transport Policy. Elsevier, 5(4), pp. 587–595.

Leonardi, J., Browne, M. and Allen, J. (2012) ‘Before-after assessment of a logistics trial with clean urban freight vehicles: A case study in London’, Procedia -Social and Behavioral Sciences, 39, pp. 146–157.

Luman, R. (2015) Stedelijke distributie in het winkellandschap van de toekomst.

Mabe, J. (2016). Micro Hubs and Shared Mobility: A Share Economy Concept for Same-Day

Delivery, available at:

http://www.techgistics.net/blog/micro-hubs-and-shared-mobility-logistics-share-economy-lastmile-sameday. [Accessed 20 November 2017].

McKinnon, A. C. and Ge, Y. (2009) ‘The potential for reducing empty running by trucks: a retrospective analysis’, International Journal of Physical Distribution & Logistics Management International Journal of Physical Distribution & Logistics Management Business Process Management Journal Iss Journal, 36(5), pp. 391–410.

McLeod, F., Cherrett, T. and Song, L. (2006) ‘Transport impacts of local collection/delivery points’, International Journal of Logistics Research and Applications, 9(3), pp. 307–317. Morganti, E., Dablanc, L. and 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, pp. 23–31.