A semi-systematic literature review, identifying research opportunities for more sustainable, receiver-led inbound urban logistics flows to large higher education institutions

The use of fully or partial systematic literature reviews in operations management, transport and logistics research has increased in recent years [8, 56, 97, 105, 121].

The Critical Appraisal Skills Program (CASP) has developed methodologies for systematic and critical reviews of literature since 1993 [35]. With roots in UK healthcare it is not tied to any discipline and has been deployed for qualitative as well as quantitative sources [34]. The approach, which acts as an inclusion/exclusion protocol is viewed as rigorous but flexible and has in recent years been adopted by operations and transport researchers in both systematic and semi-systematic reviews [33].

It has been noted that a weakness of literature reviews in the field of transport, was the lack of explicit methodology [130]. To mitigate this, I adopted a “semi-systematic review”. This is a review guided by a systematic review in terms of literature survey and selection ([31], pp. 98–112), combined with “framing a written discourse about the literature which may be established as a component part of a thesis or other research” [30].

This allowed an emphasis on such features as transparency about searching, and the potential for comprehensiveness. These “systematic” protocols, CASP for this review have been evidenced in the use explicit questions replicable research search procedures.

The narrative phase of the review allowed for a “plurality of knowing [40]” and insight from reading and understanding This addresses the suggestions that a metric driven meta-analysis is not appropriate for transport-based literature reviews, due to the “fact that many variables influence an independent variable (e.g. travel behaviour) in a complex way, resulting in complex causal relationships, and a multitude of data analysis methods and interpretations” [130].

I finalised the review with a framing using a socio-technical approach of ‘why, who, with what, where and when and how’ (5 W + h) which localised problems and solutions [135, 140] into an ensemble of actors, territory and techniques, chosen as specifically developed for this field.

The protocol adopted are shown graphically, as a flowchart, in Fig. 1 below.

Having developed this protocol, I proceeded to delimit the first searches, scoping the review.

Explicit searches were run to report the overall range and depth of literature in this top-level field. This was completed using WEB OF SCIENCE [WOS]³ Google Scholar [GS] and the EU Transport Research and Innovation Monitoring and Information System [TRIMIS⁴] databases.

A large proportion of non-peer reviewed literature existed in this field, often from research projects in the form of reports and other deliverables. Bryman called this ‘grey’ literature - of great value in many fields of research into people and systems. To enable a semi-systematic review of such materials, I utilised research outputs catalogued within the NOVELOG toolkit⁵ [135]. Since research and innovation projects and interventions had preceded academic publication in this field, this database provided a comprehensive but parallel inventory for the grey literature review process.

The field in which the review was carried out, and renewed many times, lay in the enquiry into and about the movement of freight goods and derived vehicle demand in cities. Initial reading and personal insight had revealed this to be variously defined as ‘urban logistics’, ‘city logistics’, and also ‘urban freight’. Searching for (“urban freight” or “urban logistics” or “city logistics”) in the WOS reported 1248 records as at 30/11/2018.⁶ The papers were published in fields largely related to transport, management, urban studies and environmental studies. It was apparent that this field was novel, and, over recent years, the number of publication and citations had risen sharply: suggestive of a field of research emergent in academic research.

Publication dates offered further evidence that this was a newly emergent area of publication, with 77% of all articles published after 2011, 59% of the publications were conference proceedings and 41% were peer-reviewed articles. These were followed up by a focused series of searches.

Focused searching followed using keywords and search terms developed further from the core research questions and the top-level scoping. This subset of the literature was then examined against an inclusion/exclusion list. These focused searches formed the body of work for the next step, which was a narrative review of the key literature. These focused searches covering the timespan 1970–2018. This approach was applied to each of the topics identified above, as shown in Table 1 below:

C-ITS is the utilisation of Intelligent Transport Systems (ITS) and new Information and Communication Technology (ICT) apps and infrastructures to leverage vehicle to vehicle, vehicle to infrastructure, and vehicle to human communications [110]. The theme of urban freight and C-ITS was emergent in the literature, but immature compared to the literature on C-ITS for private and public transport.

The literature had a lot of Japanese research outputs on timed delivery-window modelling, associated primarily with the research team around Taniguchi [119, 120]. That work was associated with the repurposing of traffic and network modelling at a very computational level. Taniguchi was one of the pioneers of city logistics research and was the highest output author in the top-level scoping.

Logistics sprawl was seen by researchers as the relocation of logistics facilities away from inner urban areas to suburban areas and had received an increasing level of attention from both academics and policy makers. The potential disbenefits of extending the last mile operation of an urban delivery route was exacerbated by sprawl and was inter-related to how land use planning and zoning of use varied between urban areas. Dablanc was the key foundational researcher in this field, working in both North America and Europe [47, 48]. Her work had then led to a growing body of quantitative and geo-spatial analysis from others [8, 63].

The modelling of distribution hub locations, and the balance between too close and too far away, was key to much grey and peer-reviewed literature; this had resulted in work - often led by Leonardi et al. - on the analysis, evaluation and optimisation of the break-point between the initial ‘stem’ delivery leg to a city and the ‘leaf’ delivery leg of routing [11, 68, 77, 82].

A recurring concept was that a freight landscape was peculiar to each location and time [9, 79, 80]. Rodrigue et al. had noted that “Results from New York, Los Angeles, Paris and Seoul reveal substantial variations between metropolitan areas, which are observed across the respective levels of zonal specialization as well as density changes over distance from central areas” [107]. Giuliano et al. found that, when trialled in Los Angeles, using proxies to describe the metropolitan freight landscape was of utility [57].

Land use and zoning establishment surveys were key for North American research reported in the literature. Considered to be highly dependent on the data collected in a nation or city, these tended to yield useful analysis in the USA, but were problematic in the UK and elsewhere in Europe [66, 68] due to far less stringent zoning and less rich data collection or homogeneity. This was related to freight trip generation and/or attraction, in that the attraction of freight by receivers may have been more relevant to urban logistics than traditional trip generation models [3, 66, 127].

The multi-stakeholder nature of urban freight, and the concomitant need for co-operation and collaboration, was recurrent: “A key characteristic of it is the heterogeneity of the stakeholders involved. Besides the traditional logistics actors such as shippers, carriers and receivers that share consistent interests (i.e. price and quality), city logistics highly respect the interests of public administrators and citizens that care more about the social welfare. To reach an optimal balance between private and public benefit, it is necessary to understand and in turn forecast the behaviour pattern of different groups” [15].

This need for co-operation led to the development and deployment of Freight Quality Partnerships in the UK [9, 42], and in Sweden [80], and was related to similar co-operation platforms such as “Marchandises en Ville” in France [44] and the Dutch “Platform Stedelijke Distributie” [61]. This concept of Freight Quality Partnerships - in which all key stakeholders in freight transport in a city or transport corridor met collectively to mutually solve problems, although not widely deployed outside of these examples - was emergent across the literature over time [72, 73].

A clear theme of the literature was that city logistics and urban freight was “difficult to organize, difficult to modernize” [45, 95]. This was due to local peculiarity in both place and time; the disconnect between disbenefits and the paying client; and the disruption of efficient supply chains, with the loss of time and increase in costs.

Urban Consolidation Centres (UCC) were a key theme of research and intervention. An Urban Consolidation Centre were defined as a logistics facility that was situated in relatively close proximity to the area that it served. Goods destined for this area were dropped off at the UCC and sorted and consolidated onto goods vehicles (sometimes low emission vehicles) for delivery to final destination. This was clearly differentiated from an Urban Distribution Centre (UDC), which had a wider definition and included all typical commercial logistics hubs in urban locations [10, 114].

One of the first explorations of this concept in the literature was Schuster, from Traffic Quarterly [111]. The concept had become highly popular in the 1990s, in Germany, as part of a drive for an integrated ‘city logistics’ top-down approach to freight traffic management, by local government; over 80 such schemes were noted in the trade magazine ‘Logistik Heute’, in mid-decade. All were eventually cancelled, usually due to financial non-viability [53, 54].

The widescale failure of UCCs in Germany had not prevented the concept being promoted at EU level, with multiple schemes (e.g. La Rochelle; Bristol-Broadmead) funded by EU research and development programmes, inter-regional funding etc. [74, 132]. Almost all of these had proved unsuccessful due to three key issues summarised by Martinez et al.: [lack of] Long-term financial viability; Poor selection of location; and Controlled environments [89].

Zunder [132] went further and contended that “most UCC initiatives fail in a liberal economy where free choice and market economics apply”. Despite a list of failed schemes, a few pilots - subsidized by EU or local funding – were held up as beacons to the viability of the UCC [132].

These could be divided into three key types:

There had also been trials and discussions of non-traditional approaches to consolidation, such as mobile consolidation centres [125], or virtual consolidation centres (where the consolidation was achieved through pre-planned procurement rather than any physical location [126]), or the use of crowd logistics [32].

At many points in the review there were many comments about poor ‘ex ante and ex post’ data collection [60, 75] and the poor quality and homogeneity of data standards across nations and cities [11, 14]. Worthy of note was that circa one third of all UCCs originally identified in 2005 had left no discernible data for analysis [27].

There was a significant body of work researching, analysing and modelling the use of rail, light rail, or trams, for freight delivery in urban areas. This work explored technical, operational and timetabling issues. The potential for such modal shift of freight to urban rail systems had divided opinion, require existing rail infrastructure and, would need significant infrastructure investment. Examples included the Dresden cargo tram, the Zurich recycling tram, and the short-lived Amsterdam cargo tram. The Dresden example was probably an extended ‘conveyor belt’ between separated assembly lines at Volkswagen; the Zurich recycling tram was a novel use of infrastructure for regular but non time-critical large item recycling; and the Amsterdam example failed due to the lack of a viable business model [19, 22, 29, 81, 94, 99, 106, 113].

Little of the peer-reviewed literature was on the use of EVs in freight, although a tranche of EU funded projects launched from 2012 provided grey literature on this theme, such as the FREILOT and Smartfusion projects [58, 76, 83]. Lebeau et al. developed a ‘total cost of ownership’ model to assess the competitiveness of quadricycles and light commercial vehicles for freight transport companies [75]. Herron and Coleman evaluated the volumes of EV chargers required for a certain volume of vehicles [65]. Leonardi et al. delivered many cost-benefit analyses of urban logistics schemes, using diesel, electric and cycle vehicle solutions, as did the STRAIGHTSOL project [18, 77]. For those looking for a recent review of the literature on novel alternative propulsion I suggest Oliveira et al. [97] which fell outside the scoping of this piece of research.

The use of bicycles or e-bikes for cargo delivery was well represented in the literature [62, 93]. Of note were issues of safely sharing the roads with freight vehicles [17, 100]; health and social insurance issues [87]; scheduling into existing logistics networks [17]; and the opportunity for services such as Gnewt cycles to replace the last mile in London [12]. This also related to the use of walking as a delivery mode, which was researched quantitatively in London and shown to constitute circa 62% of inner London delivery rounds for parcel carriers [12].

There was an emergent concept of the proposal and development of typologies, taxonomies, and frameworks [16, 23, 49]. This was evidenced early with the socio-technical approach of ‘why, who, with what, where and when’ which localised problems and solutions [140] into an ensemble of actors, territory and techniques. This was due to local peculiarity in both place and time, the disconnect between disbenefits and the paying client, and the disruption of efficient supply chains with the loss of time and increase in costs.

A significant development was the ‘4 As’ concept of ‘Awareness, Avoidance, Act and shift, and Anticipation’ of new technologies, by Macharis and Kin [85]. An alternative approach was taken in pragmatically repurposing taxonomies and typologies from research and EU initiatives into a novel, multi-dimensional, poly-parametric typology for city logistics, which had multiple uses in analysing and selecting interventions [135].

Recognition of the need for viable business models was evidenced, to allow transition between subsidised pilot trials to standalone viable solutions, as was a growing interest in the use of different business model frameworks and the financing of urban logistics from public-private initiatives [PPI] [24, 59, 74, 78, 84, 101, 102, 109].

The potential to deliver goods ‘out of hours’ or as ‘night delivery’ was explored differently in the literature. In Manhattan, NY there were extensive trials of receiver-led early morning deliveries, offsetting higher labour costs in office staff against reductions in congestion costs [67]. In the Netherlands, the PIEK programme had taken a strongly mechanical engineering-led approach, re-engineering vehicles and handling equipment to prevent ‘peak noise’. To comply with the standard, each product was acoustically measured and had to function emitting under 60 dB at 7.5 m from the sound source; it was then deemed suitable for out-of-hours delivery that would not cause noise disturbance to nearby residents.

A key theme of import in the review was the use of delivery and servicing plans (DSP), an establishment-led approach to analysis of receiver-led demand, developed in the Trailblazer project [127], deployed by TfL in their Palestra offices [122], and then developed further in the Smartfusion and STRAIGHTSOL projects [6,101, 117, 138]. The DSP methodology formed part of the analyses that underlay the Southampton UHS consolidation centre initiatives [78, 89].

Further to DSPs the literature held very little on urban freight and procurement activity. There was a theme of emergent ideas and proposals for mandatory municipal freight consolidation in Swedish cities and the potential power of procurement to change inbound urban freight [5, 21, 25, 37, 103, 134]. Finally, in 2018, the Transport Catapult had recommended to the UK government that “Government and Local Authorities should ...[adopt]... measures such as retiming deliveries and collaborative procurement” [89].

There was little about exploring private purchasing behaviour, save for Cherrett et al. looking at student purchases and Aditjandra et al. looking at private purchasing by University staff [36, 134].

There was a small and emergent literature on HEIs and procurement behaviour, with or without reference to freight. The work in the literature was largely written by myself and colleagues in parallel with our wider interventions, and in collaboration with colleagues from the Amsterdam Academy of Sciences, as well as a top-level overview with the Universities of Southampton and others [3, 4, 21, 90, 104, 134, 139].

There was little discussion of logistics and sustainable procurement in the mainstream procurement journals [129], sustainable procurement and supply chain logistics, from Murphy and Poist [96]. This stated that procurement could “influence the behaviour of private sector organisations” to achieve “social, environmental (and other) benefits”. However, with regard to organisational barriers, they noted unwillingness to accept higher invoiced costs as the key barrier to adoption [128]. Also of relevance was the action research work of Meehan, Ludbrook and Mason, on institutional explanations of legitimised resistance, and how “strategic avoidance responses” such as “symbolic tick-boxing” could “entrench operational barriers” [92].

Little or no theory development was evident in the literature. With the possible exception of ‘logistics sprawl’, the field was largely empirical observation with a potential tendency to ‘naïve empiricism’ [92]. The field of logistics research as a whole was weak in the late 20th and early 21st centuries in two areas: theory and explicit discussion, and statement of research philosophies [2, 26, 115]. The inability to address ‘meta’ issues may explain the difficulty in both developing theory and also “that much of the debate and criticism over methodology involves researchers who are failing to communicate with one another because they hold varying basic assumptions about their subject” [88]. The gap was a need to develop theory and, given the import of locality noted in urban freight, probably theory from case study research [50].

From these groupings of concepts I made a provisional framing using the 5 W + H approach of Zunder & Dellinger [140].