Transportation Research Record 2019, Vol. 2673(9) 502–513 Ó National Academy of Sciences: Transportation Research Board 2019 Article reuse guidelines:
sagepub.com/journals-permissions DOI: 10.1177/0361198119844252 journals.sagepub.com/home/trr
Challenges to Standardizing Emissions
Calculation of Logistics Hubs as Basis for
Decarbonizing Transport Chains on a
Global Scale
Kerstin Dobers
1, Verena Charlotte Ehrler
2, Igor Y. Davydenko
3, David
Ru
¨ diger
1, and Uwe Clausen
1Abstract
Carbon footprinting is regarded as one means of enhancing transparency on where greenhouse gas (GHG) emissions are produced within a transport chain and thus limiting the emissions and improving the efficiency of transport and transhipment on both the demand and supply sides of the market. To meet global GHG reduction targets and to realize sustainable trans-port chains, standardization of emissions calculations is progressing swiftly. One of the core requirements in the next steps of the standardization effort is harmonization of level of detail of the various transport modes. In particular logistics hubs have been identified as relevant for such a development and whereas developments for transport modes such as rail, road, air, or water are pursued by industry representatives and their organizations, logistics hubs have no organization which is intrinsi-cally motivated to further develop an approach to emissions calculation. Research can deliver an important and valuable con-tribution here. Therefore, based on extensive empirical research in the form of questionnaires and real-life examples of emissions calculation, this paper describes the motivations and barriers currently experienced by shippers and logistics ser-vice providers when computing emissions. Possible approaches to overcoming these barriers and contributing to the further improvement of the level of maturity of emissions calculation of logistics hubs are described and discussed. The paper closes with an outlook on further requirements toward transport chain emissions calculation standardization developments.
The United Nations Framework Convention on Climate Change in Paris, COP21, set the target of keeping global warming below 2°C. To reach this target, the Kyoto con-ference greenhouse gas (GHG) emissions have to be reduced by around 13.5 Gt CO2e by 2030 (1). Transport
and logistics activities (i.e., passenger and freight trans-port) contribute substantially to the global stock of GHG emissions. They accounted for around 7.5 Gt CO2 in
2015, which is 18% of global emissions by human activity (2), and will be one of the economic sectors most difficult to decarbonize (3–5). Almost 10 years ago, the World Economic Forum estimated that logistics (transport, warehouses, and terminals) accounted for around 5.5% of global GHG emissions (6), 13% of which were allo-cated to ‘‘logistics buildings’’ (4). As McKinnon recently observed, ‘‘very little data is available on GHG emissions from the buildings and terminals in which goods are stored, handled and transshipped’’ (4, p. 14). He further cited estimates on the contribution of warehouses to transport emissions, according to which warehouses gen-erate approximately 20% of the transport emissions in
the U.S.A. (7), and approximately 11% in the U.K. (8). For Germany, a share of around 15% was estimated for logistics nodes (i.e., warehouses, terminals, and ports) (9). The statistics on GHG emissions by transport and logistics nodes underline that the efficiency of current transport chains has to be improved and their emissions reduced. This requires, however, further insight into cur-rent emission levels and the development of an interna-tionally accepted standard for the calculation of transport chain emissions to support comparisons, and with them the identification of potentials for reducing
1
Fraunhofer Institute for Material Flow and Logistics IML, Dortmund, Germany
2
German Aerospace Center (Deutsches Zentrum fu¨r Luft- und Raumfahrt, DLR), Institute of Transport Research, Berlin, Germany
3
Netherlands Organization for Applied Scientific Research TNO, The Hague, The Netherlands
Corresponding Author:
Address correspondence to Kerstin Dobers: [email protected]
GHG emissions and best practice. During recent years, distinct progress towards the establishment and imple-mentation of an international standard for the calcula-tion of transport chain emissions has been made by the Global Logistics Emissions Council (GLEC) covering research, industry, and standardization bodies as well as other stakeholder organizations. Starting from existing base methodologies (e.g., EN 16258 (10), SmartWay (11), IMO EEOI (12), GHG protocol (13)), the GLEC Framework (14) was published, an internationally devel-oped and applicable methodological framework for the calculation of transport chain emissions, whose update is advertised for June 2019. One formalized global stan-dard, the ISO norm, covering all modes on a comparable level, supporting equally ex-post as well as ex-ante calcu-lations, and facilitating unambiguous data collection and exchange, is not yet implemented.
Several industry organizations, shippers, and logistics providers as well as their representatives work on further developing guidelines and principles for emissions calcu-lations for the transport modes they are working with. Guidelines and frameworks for air, road, sea, inland waterways, and rail are therefore in progress. Logistics hubs are a weak link as far as a harmonization of the various levels of carbon footprinting maturity of the indi-vidual transport modes is concerned, despite their impor-tance within the transport chain. Research can deliver an important contribution here by progressing the approach to computation of the carbon footprint of logistics hubs.
Therefore, it is the aim of this paper to focus on logis-tics hubs and their particular challenges. To ensure prac-ticability of the emissions calculation method, empirical research has been conducted, including extensive inter-views with industry partners and the analysis of real-life emission calculation pilot cases of these industry part-ners. This empirical research forms a basis for the meth-odological suggestions developed in this paper.
The paper is structured as follows. Empirical research revealing motivations, expectations, and barriers cur-rently experienced by the industry when calculating their transport chain emissions is presented in the next section. Based on these findings, a possible approach for captur-ing and computcaptur-ing emissions of logistics hubs is pre-sented and discussed in the third section. The paper closes with an outlook on further requirements toward the development of standardization of transport chain emissions calculations.
Drivers and Barriers to Emissions
Standardization Efforts—Industry
Perspective
During the period from summer 2017 to summer 2018 extensive empirical research (15) was carried out to
identify industry’s perspective on drivers, motivators, and barriers, as well as practical implementation issues for the advancement of a standardized method for the calculation of transport chain emissions. Using a uni-form questionnaire, a selection of 30 industry partners were interviewed on their motivations, their expectations, as well as their experiences to date, related to calculation of the emissions of their transport chains. The industry partners included internationally operating shippers, freight forwarders, carriers, and logistics service provi-ders, with company sizes ranging from SMEs to multina-tionals, from Europe, as well as the Middle East, Asia, and North America. They covered all transport modes and used or operated logistics hubs. As well as the questionnaire-based interviews, real-life test cases of the industries’ operations were used to calculate emissions using the GLEC Framework (14). In the following, the findings of this empirical research are summarized, describing the drivers and motivations of the industry partners interested in the calculation of their transport chain emissions, as well as currently experienced chal-lenges and obstacles to successfully computing emissions in their everyday operations.
Drivers for GHG Emissions Calculation and Accounting
in Supply Chains
The drivers for emissions calculation and accounting in transport chains can be divided into three broad cate-gories. The first category is related to the corporate sus-tainability programs (CSP), which can often be translated into a set of sustainability performance and improvement targets. The underlying reasons for CSPs are a corporation’s value system and strategy, the man-agement’s belief in the corporation’s responsibility toward the environment, as well as the corporation’s value system displayed toward customers and third par-ties. These CSPs often result in the adoption and display of environmental key performance indicators (KPI) and targets as part of an organization’s reporting, for exam-ple, in its annual report. The second category is commer-cial reasons, namely the ability to satisfy contractual requirements, where measuring GHG emissions of trans-port and logistics activity may be one of the tendering requirements. The third category is anticipatory, related to expectations by the business with respect to manda-tory emissions measurements and reporting, as well as expectations with respect to market conditions.
CSPs often define specific targets for emissions reduc-tions over a specific period. For instance, one large ship-per has a target of 40% reduction in GHG emissions in its transport and logistics activities by 2020 compared with the base year of 2010. Such companies may have a position dedicated to the monitoring and realization of
such a target (e.g., chief environmental officer). These specialists are directly accountable for ensuring adher-ence to the set emission targets, in cooperation with the organization’s various departments. If environmental reporting is incorporated into the annual reporting, the company’s accountants may set strict criteria on how the emissions are computed and accounted, for instance, mandating the use of certain emission factors, or pre-scribing the use of assured information provided by the organization’s suppliers, in particular the logistics service providers (LSPs).
Consequently, for LSPs the driving force behind emis-sions calculation and emisemis-sions accounting are their own CSPs and the commercial reason to satisfy customers’ requests for provision of the related data. As default fac-tors do not fully reflect a company’s sustainability efforts, for example, increased load factors and resulting fewer transport movements and lower emissions, usually measured data is preferred by both LSPs and their cus-tomers. In fact, tendering processes of shippers may favor those LSPs that are capable of providing real-world emission data, specifically related to the transport and logistics operations of the shipper. Furthermore, GHG emissions are closely related to the fuel type and fuel use, therefore calculation of the emissions helps LSPs to improve their own performance, as well as pro-viding a quantitative KPI to monitor and control.
Furthermore, by now many companies anticipate that government policies related to the GHG emissions of transport and logistics activities will become stricter. Specifically, national and supranational (e.g., European Commission) governments may in the future mandate application of a GHG emissions measurement and calcu-lation methodology for the purpose of mandatory emis-sions reporting (e.g., 16, 17). This anticipation leads to a willingness to show that the sector takes the issue seri-ously and is prepared for a possible change in legislation. Additionally, this results in organizations taking action and contributing directly to the efforts on standardiza-tion of emissions calculastandardiza-tions. This behavior enables industry to contribute actively to the development and definition of such a standard providing a basis for the future legislative and regulatory developments.
Main Barriers to Be Overcome in Emissions
Calculations and Accounting in Transport and Logistics
This section identifies and provides a short explanation for the following six classes of barriers of emissions cal-culation and accounting in transport and logistics. Motivational Barriers (Demand and Use). Not all businesses that provide or use transport and logistics services see a need for computation of GHG emissions related to theseactivities. This is mainly because of not perceiving a ‘‘busi-ness case’’ for the effort, besides reluctance to share the data. At present there are various stakeholders with an interest in understanding general emission sources related to transport and logistics activities, but there appears to be little demand for accurate computation of emissions. Therefore, investment in such a capability often does not present a viable business case. Furthermore, as fuel is one of the most important expenses for LSPs, calculating emis-sions based on real operational data may reveal the cost structure of the LSP’s operations and weaken the com-pany’s negotiating position, or disclose essential opera-tional data to its competitors. Moreover, there are no well-known cases on the use of emissions data beyond CSP programs, as opposed to an ideal outcome where these data are used to optimize transport and logistics chains with respect to resulting GHG emissions.
Methodological Barriers. There are still some methodologi-cal gaps present, such as computation of GHG emissions for logistics sites, in some special cases of transport, such as road haulage of air cargo, to name a couple of exam-ples. There are still ongoing efforts to close the gaps and harmonize different approaches, making them easier to understand and implement. At present, a seemingly frag-mented landscape of various programs with different methodological approaches may be partly overcome by means of cross-accreditation schemes and a push towards standardization. A standard on emissions calculation, assurance, and reporting would greatly alleviate the problem of diversity of methodological approaches. The authors deem an ISO standard to be the best outcome for harmonization of methodological solutions.
Information Asymmetry. Shipment level emissions computa-tions based on real-world information needs two classes of data. The first class of data is the relevant mass of GHG emissions, which is proportional to the amount of fuel used. The second class of data is the transport activ-ity, measured as weight (or volume) shipped over a dis-tance: the measure most often used is tonne-kilometers. The transport providers generally have the data on vol-ume of fuel used by their vehicles; the users of transport do not have this information if transport is carried out by a third party LSP, however, they generally have some good data on their transport activity. The challenge is to overcome the fact that different parties in the chain pos-sess only parts of the necessary data.
Implementation. Implementing carbon footprinting and carbon accounting requires data gathering and realiza-tion of proper computarealiza-tion. This process may be costly because of the data collection effort required to gather
the data within the company, including the possibility of using data from other parties. The implementation can require a number of iterations to debug flaws and errors.
Data Exchange. Emission data exchange between organi-zations (e.g., between LSPs and shippers) has to overcome technical implementation barriers and privacy concerns. The privacy concerns involve the non-disclosure of com-mercially sensitive data and likewise the methodological concern that shared emissions data does not reveal too much sensitive information that might lead to the weak-ening of a party’s negotiation position.
Default Emission Factors. In many cases no complete real-world data are available for carbon footprint calculation. In such cases, it is possible to use aggregated average data (default factors) on emissions to substitute missing data, though most methodologies downgrade the output of such emissions computations (e.g., from a ‘‘gold’’ level to a ‘‘silver’’ level). There is a tradeoff on granularity of the default emission factors: too general default emission factors do not sufficiently capture the specifics of opera-tions, while very detailed emission factors require knowl-edge on the scope of applicability and data collection, and may be unavailable.
Industry Outlook and Expectations toward Future
Developments
The major barriers experienced by industry at present are related to the collection and exchange of data, the guidance and access to default data for cases where no measured data is available, and methodological barriers. Therefore, one focal industry expectation toward future developments is related to the methodological maturity of a harmonized approach to emissions calculation related to the operation of various modes, in particular logistics hubs (see Figure 1).
As far as methodological barriers are concerned, the current different levels of standardization efforts of the various transport modes and the lack of guidelines for capturing logistics hubs emissions are identified as the most pressing issues. Industry and industry related orga-nizations are progressing the development of emissions calculation approaches in particular for those modes which are most relevant for them. Consequently, devel-opments not allocated to one specific mode are often not pursued with the same support. This paper aims to con-tribute to closing this gap and in the following focuses on the description and discussion of possible approaches to emission accounting at logistics hubs as these are con-sidered relevant next developments, given their impor-tant role within transport chains.
Industrial IT platforms for emissions calculation at the level of intermediary between the shippers and their LSPs can potentially alleviate two problems, namely sen-sitivity with respect to sharing of confidential operational data and the complexity of sharing information in a many-to-many organizational relationship. For instance, an LSP can serve a large number of clients, while a large shipper may use the services of hundreds of LSPs. Even if emission data were available on the part of LSPs, sharing and receiving the data would be a laborious endeavor. An IT platform can be a place that gathers emission-related data of LSPs, such that for LSP it is a one-to-one connection; the shipper would also have a one-to-one connection to such a platform for acquiring all the neces-sary emission data related to its transport and logistics operations. Furthermore, IT platforms may potentially be useful for overcoming the problem of the various methodological approaches, thus allowing co-existence of different programs and possibly standards on carbon accounting through generation of carbon footprint and carbon accounting data using different methodological approaches.
Approach for Emissions Accounting at
Logistics Hubs at an Activity Level
Among the list of barriers identified by the empirical research is the issue of methodological clarity with respect to logistics hubs. This section prioritizes the development of a harmonized methodology for logistics hubs over other barriers and provides deeper methodolo-gical suggestions on how to tackle logistics hub emissions in a uniform way with the goal of incorporation of logis-tics hubs emissions into an emission computation stan-dard for the transport and logistics sector.
The status-quo analysis within the empirical research also focused on logistics hubs referring to all sites ‘‘that combine different transport legs (within and between modes) or are the starting or end point of transport chains. As such, transhipment centers include terminals at sea or inland ports, freight and intermodal terminal, terminals at airports, warehouses, cross-docking sites, or distribution centers etc.’’ (19). Most progress can be stated for the calculation of emissions from container terminals at sea ports (20, 21), most recently by Spengler and Wilmsmeier (22). Before January 2019, there was a lack of a commonly accepted approach in the field of warehouses and transhipment sites. The most relevant gap related to logistics sites was the lack of a definition of transhipment center categories and services. ‘‘The variety of logistics nodes is vast and their functions and function-alities are very different, resulting in completely different energy use and emissions caused by them. Moreover, gui-dance for the capturing of non-energy use related
emissions such as e.g. refill of refrigerants that were lost e.g. during transport, wrapping material or waste pro-duced is still missing’’ (23, p. 5). In early 2019 the ‘‘Guide for GHG Emissions Accounting for Logistics Sites’’ (24) was published, the underlying approach of which is sum-marized in the following.
Methodology and Calculation Approach
The objective for GHG emissions accounting at logistics sites is to cover all relevant emission drivers of the activi-ties at the sites and to derive representative and meaning-ful emissions indicators for monitoring and reporting within a company. This helps to review expected improvements of measures at site level as well as along the transport chain. In general, emissions indicators are meaningful for companies to identify and implement optimal green levers to meet given GHG emissions reduction targets.
Types and Functions of Logistics Centers. Considering the wide variety of logistics sites and their even wider range of operations and services, it is necessary first to develop a clear classification scheme for emissions accounting. Here, the characteristics that can be used for the distinction of logistics services might cover aspects such as:
Type of cargo handled (e.g., containerized, rolling stock, bulk, and liquid).
Modes connected to the site (e.g., is it a terminal at a seaport or an inland port?).
Whether operations and equipment are outdoor (e.g., terminal) or indoor (e.g., warehouse). Emission sources to be covered (scope 1, 2, and 3
according to the GHG protocol [13]).
Relevant additional (value-added) services offered at the site and to be covered by the assessment (e.g., services to drivers, cargo, and workers, i.e., gas stations, workshops, warehouses).
Higgins et al. (25) developed a classification of logistics sites based on functionality and value added services, covering: (1st level) warehouse and distributions centers, (2nd level) freight distribution centers, and (3rd level) gateway clusters. This was further detailed by Ru¨diger et al. (26) using the cri-terion of contracts to distinguish different logistics facilities. This classification scheme defines transhipment sites, ware-houses, and distribution centers as follows:
Transhipment sites: Only contracts in regard to transhipment of goods with and/or without specifi-cations on temperature requirements are relevant. Warehouses: Only contracts on ambient and/or
refrigerated storage performance, for example, in regard to throughput, stock levels, numbers of required storage locations and storage duration, and min/max temperature requirements; in rela-tion to contract logistics with addirela-tional specifica-tions for order picking processes (e.g., packaging requirements) are relevant.
Distribution centers: Any combination of con-tracts in regard to transhipment and warehousing activities (as described above) with and/or without specific temperature requirements.
Further specialized logistics sites such as re-packing sites, for example, in the automotive industry or parcel centers used within courier, express, and parcel (CEP) networks can be allocated to this scheme although the operators may use different terms.
The operations at logistics sites (see Figure 2) cover the receipt of goods, inbound handling, buffering or stor-age, outbound handling, and order picking as well as dis-patch. To carry out these operations it is usually necessary to have lighting (indoor and outdoor), to fulfill temperature requirements (heating or refrigerating), and to implement back office tasks.
Environmental Key Performance Indicators for Logistics Sites. To monitor the GHG emissions performance of logistics sites over the years and to identify the effect of green measures (e.g., installation of LED lighting), it is neces-sary to calculate GHG emissions intensity values. These indicators outline the increase (or decrease) of efficiency of services or processes. The term ‘‘intensity’’ refers to the resources used (e.g., energy consumption) and the associated emissions per unit of desired output (e.g., per throughput or per defined activity at the logistics site) (28).
In the context of logistics sites the resource consump-tion caused by operaconsump-tions and emissions emitted have already been analyzed and described (e.g., 24, 26, 29, 30). Referring to this, the authors summarize the recom-mended assessment boundaries approach as follows. Aligned with the life-cycle approach for energy use as proposed by the GHG protocol (13) and the GLEC Framework (14), the total fuel and electricity consump-tion of all relevant operaconsump-tions (see Figure 2) are assessed. Adding the leakage of refrigerants at sites with
temperature controlled conditions, the assessment boundaries of calculating GHG emissions according to the GHG protocol (13, 31) are:
Scope 1 emissions (burning of fuels, leakage of refrigerants)
Scope 2 emissions (purchased electricity, steam, heating, and cooling for own use)
Category 3 of upstream scope 3 emissions (fuel-and energy-related activities, not included in scope 1 or scope 2)
The annual emissions of a logistics site can be divided by the annual throughput (e.g., total amount of tonnes or pallets outbound) to calculate an average emission or carbon intensity value for all goods leaving the site. This helps companies to gain an initial idea of a change in efficiency of the site from one year to another. However, to identify green measures at equipment level, more detailed environmental indicators are usually required. To prepare the derivation of more detailed environmen-tal indicators, the three classified types of logistics sites (transhipment sites, warehouses, and distribution cen-ters) are further separated to activity level.
Table 1 shows the allocation of the resource consump-tion to operaconsump-tions and activities at logistics sites. According to this, all listed operations and activities may be electrified. However, the handling of units can also be realized by equipment (e.g., truck conveyors, fork lifts, or rack feeders) run with diesel, petrol, gas, or hydrogen. Thus, the goods may be handled by different equipment and, respectively, consume different energy sources depending on the operations required for the relevant activity. Moreover, depending on the climate zone of the logistics sites as well as the goods handled, a continuous temperature control of the indoor logistical area is neces-sary, which may be refrigerating, heating, or both. For this, additional energy sources as well as the refill of leaked refrigerants have to be considered.
Against this background, activity categories are derived for logistics sites differentiating three main per-spectives (also see Table 2):
Temporal requirements: Transhipment and/or storage of goods. At transhipment sites goods are often buffered for better transport bundling, as such goods leave the site typically after 24 h at most.
Temperature requirements: Ambient and/or refri-gerated goods. The differentiation of only two temperature levels is a simplification to gain greater transparency of the methodology and applicability at this point. For some sites it may be reasonable to differentiate further between
different temperature levels (e.g., ambient, fresh, cooled, and frozen goods) for receiving more meaningful, activity-related, emission intensity values.
Picking requirements: Goods leave the site with or without being order picked. Shipments are broken down into smaller shipments and/or orders are picked for final delivery.
By this, in total eight different activity categories are derived for differentiating goods leaving logistics sites. For transhipment sites, warehouses, or distribution cen-ters with only one temperature zone, this number reduces to a maximum of four activity categories, since some of the eight can be ignored accordingly.
The total GHG emissions of the logistics site (EMtotal) are calculated by multiplying the resource consumption (Qi) one by one by the resource specific (i) emission fac-tor (EFi) and adding them all up according to:
EMtotal= X
Qi3 EFi= EMheat+ EMrefrigerating
+ EMstorage+ EMpicking+ EMgeneral ð1Þ
Emission factors are used to convert energy used or refrig-erants leaked into GHG emissions. Behind each emission factor are a number of general principles of GHG emis-sions accounting as well as former life cycle assessment studies which may result in a large variety of emission fac-tors for the same issues. As such, it is necessary to agree internationally on a basic set of factors so that the calcu-lated emissions inventories are comparable (24).
Furthermore, it is necessary to keep calculated partial emissions separate for the emissions assessment and allo-cation at activity level, that is, partial emissions for heat-ing, for refrigeratheat-ing, for storage, for order pickheat-ing, and for general activities (e.g., goods receipt, dispatch, office, IT systems), as shown in the formula above. Figure 3 provides an overview on the allocation procedure for a warehouse offering order-picking. Since ambient as well as refrigerated shipments are handled at the facility, the electricity used for the refrigerating equipment and asso-ciated emissions will not be allocated to ambient ship-ments, but to refrigerated ones only. The same holds true for the potential leakage of refrigerants.
The underlying principles of the allocation procedure are that in a first step the partial emissions are divided by
Table 1. Equipment Used at Logistics Sites and Their Resources Used (24)
Equipment for
Example energy
consumer Diesel/petrol Gas*/hydrogen Electricity Heating** Refrigerants *** Handling of logistics
units: (un)loading, transport on-site, storage
Industrial truck, conveyor, fork lift, high bay warehouse, rack feeder, lift, crane
O O O
Temperature control Refrigerant device O O
Heating device O O
Lighting Lighting O
Repacking, packing, order picking, sorting
Wrapping machine, sorting machine
O Others (basics) IT systems, offices,
garbage compactor, weighing equipment
O O
*For example: LNG (liquefied natural gas), CNG (compressed natural gas).
**For example: natural gas, heating oil, district heating, geothermal energy, wood chips or pellets. ***For example: R-134a, R-404A, R-407C, R-410A, R-744, R-717.
Table 2. Classification Scheme for Logistics Sites and Their Activities (24)
Requirements regarding
Time (stock-keeping) Temperature Order picking
Characteristics No storage (transhipment) Ambient above +8°C Without order picking With storage Short-term Medium-term Long-term Refrigerated Fresh ( +4°C to +7°C) Sensitive (0°C to +2°C) Pharmaceutical product ( +2°C to +8°C) Frozen (\ 0°C), in case of food \ –18°C
the relevant of quantities of logistics units for calculating allocation coefficients as follows:
ac1= EMheat Qambient units , ac2= EMstorage+ EMgeneral Qtotal units , ac3= EMrefrigerating Qrefrigerated units , and ac4= EMpicking Qpicked units ð2Þ
In a second step, the respective allocation coefficients are summed up as indicated by the puzzle pieces shown in Figure 3. Thus four different emission intensity values are calculated for the respective four activity categories of the example warehouse, that is, handling of (i) ambi-ent units with order picking, (ii) ambiambi-ent units without order picking, (iii) refrigerated units without order pick-ing, and (iv) refrigerated units with order picking.
Example
The calculation and allocation approach described for GHG emissions accounting of logistics sites has been applied for various sites in Europe. For confidentiality reasons, industry data has been further processed and anonymized for this paper to derive the following exam-ple ambient distribution center in Europe. In a represent-ing year, the site consumes 765,000 kWh electricity and 300,000 kWh natural gas for heating annually. Using smart metering for relevant consumers and processes, the operator can allocate the electricity consumption to those listed in Figure 4.
The total GHG emissions of the site account for 444 tonnes CO2e/year, which refers to an average of 8.5 kg
CO2e/tonne handled (total 52,000 tonnes/year). In this
example, the distribution center offers three main activi-ties: transhipped goods without order picking (26,000
tonnes) as well as stored goods with (15,600 tonnes) or without order picking (10,400 tonnes).
Using the allocation principles described above, emis-sion intensity values can be calculated for all three activi-ties (see Figure 3). The figure shows clearly to what extent the individual processes contribute to these values. For example, this approach supports transparently the impact of an operator’s plan to implement improvement measures to reduce the electricity consumption of storage equipment by 25% (see bars ‘‘Stored (improved).’’ in Figure 4). This holds true for other green measures such as LED lighting, the use of another heating system, or the choice of green tariffs for the electricity supply used at the distribution center. Therefore, the developed GHG emissions intensity values provide meaningful informa-tion for companies on their path of emissions reducinforma-tion.
Development of Average GHG Emission Intensity
Values for Different Center Types
As outlined in the introduction, little information is available on GHG emissions of logistics sites and their emission intensity. To overcome this gap, this paper summarizes an initial analysis of 196 European logistics sites of 42 companies, for which operators have provided annual information on, for example, energy consump-tion, refill of refrigerants, and throughput. Because of varying data availability and quality (e.g., completeness of assessment boundaries, provision of base units ‘‘tonnes,’’ ‘‘parcels,’’ or ‘‘letters’’), a reduced number of 189 sites has been used to calculate emission intensity values per site. Moreover, for some site types only data from less than three different companies could be col-lected (especially CEP sector), which reduces the further processed and published information for keeping confi-dentiality to 53 sites in total.
Figure 3. Allocation procedure for a warehouse providing emission intensities per ambient and refrigerated unit with or without order picking (24).
This data base has been used for further analysis dif-ferentiating three site types, that is, ambient tranship-ment sites as well as ambient and refrigerated sites offering both transhipment and storage. The site sizes vary from 2,800 tonnes to 6.3 million tonnes outbound with 65,000 tonnes as median value. All operators spec-ified a site-specific electricity mix, however, an average European emission factor (479 g CO2e/kWh [14]) was
used for calculating the average emission intensity value as specified in Table 3. Almost all sites use natu-ral gas as heating energy source, seven sites use heating oil; the use of district heating and geothermal or wood-based energy is rare. Refrigerated sites refilled the fol-lowing refrigerants: R-410A, R-404A, or R-134a.
Table 3 summarizes the results of the analysis covering 53 logistics sites. The values vary significantly for each type of logistics site: for example, in the case of ambient storage sites from 0.49 kg to 57.94 kg CO2e/tonne with a median
value of 7.26 kg CO2e/tonne. Considering the constrained
sample size, the authors refrain from further interpreting the data at this point. Instead, the relevance for further research tasks shall be outlined to establish useful average emission intensity values for logistics sites in the future.
Extension of Sample Size. A sound approach for calculating GHG emissions of logistics sites is now available. Starting from this, existing platforms such as
EcoTransIT World (32), Sustainable Performance Monitor (33), greenrouter (34), TK’Blue (35), REff-Tool (36) or Lean & Green (37) are now in the position of col-lecting environmental data of logistics sites against a harmonized framework (24). The authors together with GLEC are currently supporting the merging of these efforts so that a sufficient data base can be expected in due time.
Definition of Environmental Performance Indicators. The current approach for calculating emission intensity values of logistics sites uses weight metrics for providing informa-tion that can be used within logistics chain calculainforma-tions. Further research is needed to better understand whether other metrics may provide advantages for identifying green measures along logistics chains, for example, the shipment volumes, or the use of the internal floor area or volume of the building.
Establishing Average Emission Intensity Values. An extended database facilitates comprehensive research on and the development of average emission intensity values that (i) reflect regional industrial preferences for technologies and climate conditions, (ii) consider industry sectors with additional effects on emissions, for example, CEP, food, automotive, clothes, and (iii) take into account the classi-fication scheme for logistics sites (as defined in Table 2).
Figure 4. Calculated emission intensity factors at activity level of an ambient site (16).
Table 3. GHG Intensity Values for Three Types of Logistics Sites in Europe (16)
Type of logistics sites Number of sites Minimum Median Maximum
Transhipment (ambient) 4 0.07 kg CO2e/t 1.36 kg CO2e/t 2.13 kg CO2e/t
Storage + transhipment (ambient) 34 0.49 kg CO2e/t 7.26 kg CO2e/t 57.94 kg CO2e/t Storage + transhipment (refrigerated) 15 2.31 kg CO2e/t 16.25 kg CO2e/t 229.49 kg CO2e/t
In this context, further research with regard to emerging technological changes in modes (e.g., e-mobility, autono-mous vehicles) (38) or architecture of logistics buildings (e.g., multi-story warehouses) (39, 40) is required as well, as they may affect logistics site operations and location choices and, thus, consequently future emission intensity values.
Summary and Outlook
This paper aims to advance research on methodological guidelines for emissions calculation for logistics hubs. The aim of this effort is to facilitate discussion and speed up preparation of an international standard on emission accounting in transport and logistics. The described empirical research that builds the basis for the work was carried out in close cooperation with industry. Questionnaire-based interviews with industry representa-tives were complemented with pilot tests during which transport chain emissions for real-life cases were calcu-lated using the GLEC Framework (14). Findings show that barriers experienced by organizations in their attempts to calculate their transport chain emissions are mainly related to the sourcing of data, data communica-tion between the partners of the chain, and the data quality. Further barriers are related to asymmetry of detail and clarity of guidelines for emissions calculation between the various modes. Logistics hubs, because of their central role within supply chains, have been identi-fied as one of the most pressing elements for which emis-sions calculations need to be further developed to ensure the ability of emissions calculation standardization to deliver meaningful results that contribute to reducing GHG emissions of the transport and logistics sector. The paper suggests a classification approach for logistics hubs and for the activities carried out at their premises as a starting point for further specifications of emissions calculation. Furthermore, a detailed approach to captur-ing and computcaptur-ing the emissions of logistics hubs is developed, described, and discussed. Using empirical findings, the paper incorporates emissions of logistics hubs into the methodological toolset, thus emphasizing once more the relevance of measuring and understanding their efficiency and energy use.
In the next step verification and testing with the indus-try in further pilot cases is suggested, to ensure its usabil-ity and to maximize its acceptance. Following such tests, the approach should be included in ongoing formaliza-tion processes of the standardizaformaliza-tion efforts, that is, the development of an ISO norm. Beyond this methodologi-cal research, further effort has to be made to establish a data exchange format and protocol as well as a neutral platform, via which emission data can be exchanged
between transport chain partners without forcing them to share operational data with others.
As this paper has shown, emissions calculation stan-dardization efforts have come a long way, and the ongoing developments are promising steps to deliver a frame for the improvement of transport chain efficiency and, subsequently, to reduce the sector’s GHG emissions.
Acknowledgments
This research is based on work carried out within the LEARN project (Logistics Emissions Accounting & Reduction Network), co-funded by the EU within Horizon 2020 frame-work (grant number 723984), within research funded by Fraunhofer Gesellschaft as well as Smart Freight Centre/ GLEC and EcoTransIT World. The authors thank all funders as well as all companies and site operators that have been sup-porting and actively contributing to make this research possible.
Author Contributions
The authors confirm contribution to the paper as follows: study conception and design: KD, VCE, IYD, UC, DR; data collec-tion: KD, VCE, IYD, DR; analysis and interpretation of the results: KD, VCE, IYD, DR; draft manuscript preparation: KD, VCE, IYD. All authors reviewed the results and approved the final version of the manuscript.
References
1. United Nations Environment Program. Gap Report 2016. 2016. http://www.unep.org/emissionsgap/. Accessed August 1, 2018.
2. OECD and ITF. ITF Transport Outlook 2017. OECD Pub-lishing, Paris, France. 2017. http://dx.doi.org/10.1787/ 9789282108000-en.
3. Sims, R., R. Schaeffer, F. Creutzig, X. Cruz-Nu´n˜ez, M. D’Agosto, D. Dimitriu, M. J. Figueroa Meza, L. Fulton, S. Kobayashi, O. Lah, A. C. McKinnon, P. Newman, M. Ouyang, J. J. Schauer, D. Sperling, and G. Tiwari. Transport. In Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (Edenhofer, O., R. Pichs-Madruga, Y. Sokona, E. Farahani, S. Kadner, K. Seyboth, A. Adler, I. Baum, S. Brunner, P. Eickemeier, B. Kriemann, J. Savolainen, S. Schlo¨mer, C. von Stechow, T. Zwickel, and J.C. Minx, eds.), Cambridge University Press, Cam-bridge, MA, and New York, NY, 2014. http:// www.ipcc.ch/pdf/assessment-report/ar5/wg3/ipcc_wg3_ ar5_chapter8.pdf. Accessed August 1, 2018.
4. McKinnon, A. C. Decarbonizing Logistics: Distributing Goods in a Low Carbon World. Kogan Page, New York, NY, 2018.
5. Cliff, S. S., P. T. Dube, and S. Edwards. Decarbonizing the Logistics and Long-Distance Transportation of
Freight. In Decarbonizing Transport for a Sustainable Future Mitigating Impacts of the Changing Climate. Sum-mary of the Fifth EU-U.S. Transportation Research Sym-posium. 17–18/05/2017. Washington, D.C., 2018.
6. World Economic Forum. Supply Chain Decarbonization: The Role of Logistics and Transport in Reducing Supply Chain Carbon Emissions. World Economic Forum, Gen-eva, Switzerland, 2009. http://www3.weforum.org/docs/
WEF_LT_SupplyChainDecarbonization_Re-port_2009.pdf. Accessed July 30, 2018.
7. Ries, J. M., E. H. Grosse, and J. Fichtinger. Environmen-tal Impact of Warehousing: A Scenario Analysis for the United States. International Journal of Production Research, Vol. 55, No. 21, 2018, pp. 6485–6499. https: //doi.org/10.1080/00207543.2016.1211342.
8. Baker, P., and C. Marchant. Reducing the Environmental Impact of Warehousing. In Green Logistics: Improving the Environmental Sustainability of Logistics (McKinnon, A., M. Browne, M. Piecyk, and A. Whiteing, eds.) 3rd ed., Kogan Page, London, UK, 2015, pp. 194–226.
9. Ru¨diger, D., K. Dobers, V. Ehrler, and A. Lewis. Carbon Footprinting of Warehouses and Distribution Centers as Part of Road Freight Transport Chains. Proc., 4th Interna-tional Workshop on Sustainable Road Freight Transport, Cambridge, UK, 2017.
10. EN 16258:2012 Methodology for Calculation and Declara-tion of Energy ConsumpDeclara-tion and GHG Emissions of Trans-port Services (Freight and Passenger). European Standards, 2012, p. 68.
11. United States Environmental Protection Agency. Smart-Way. 2017. https://www.epa.gov/smartway. Accessed July 30, 2018.
12. Guidelines for Voluntary Use of the Ship Energy Efficiency Operational Indicator (EEOI). MEPC.1/Circ.684. Interna-tional Maritime Organization, London, UK, 2009. 13. World Resources Institute, and World Business Council for
Sustainable Development. The Greenhouse Gas Protocol: A Corporate Accounting and Reporting Standard. Washing-ton, D.C., 2004. https://ghgprotocol.org/sites/default/files/ standards/ghg-protocol-revised.pdf. Accessed July 30, 2018.
14. Greene, S., and A. Lewis. GLEC Framework for Logistics Emissions Methodologies. 2016. www.smartfreightcen-tre.org/glecframework/glecframework. Accessed July 30, 2018.
15. Logistics Emissions and Reduction Network LEARN. Testing & Validation. 2018. http://www.learnproject.net/ main/activities/testing-validation. Accessed July 30, 2018. 16. CE Delft, Fraunhofer IML, TRT, Ecorys, and Conlogic.
Fact-Finding Study in Support of the Development of an EU Strategy for Freight Transport Logistics. Lot 3: Introduction of a Standardized Carbon Footprint Methodology. European Commission DG MOVE, 2014. https://ec.europa.eu/trans-port/sites/transport/files/themes/strategies/studies/doc/ 2014-12-introduction-of-a-standardised-carbon-footprint-methodology.pdf. Accessed July 30, 2018.
17. Sexton, T., and O. Lah. The Influence of Policy Environment Factors on Climate Change Mitigation Strategies in the Transport Sector. In Decarbonizing
Transport for a Sustainable Future Mitigating Impacts of the Changing Climate. Summary of the Fifth EU-U.S. Transportation Research Symposium. 17-18/05/2017. Washington, D.C., 2018.
18. Ehrler, V., K. Dobers, and I. Davydenko. Expectations and Requirements toward Future Harmonization Develop-ments. In Research and Development Agenda Towards Eco-Labeling for Transport Chains. http://www.learnproject. net.
19. Ehrler, V., S. Seidel, A. Lischke, D. Ru¨diger, K. Dobers, A. Lewis, D. Luzzini, and S. Val. Status Quo of Research and Publications Development on Emission Calculation Stan-dardization for Transport Chains. LEARN Project Deliver-able D3.1. 2016. www.learnproject.net. Accessed July 30, 2018.
20. Capelle, S., C. Feighan, A. Woods, P. Pirrello, A. Schruth, J. H. Pietsch, and S. Mordeglia. Guidance for Greenhouse Gas Emission Footprinting for Container Terminals, 2017. http://navclimate.pianc.org/news/download/21_9cd57e0cb 26d8f27913e702868f8828e. Accessed July 30, 2018. 21. Global Maritime Energy Efficiency Partnerships GloMEEP.
2017. http://glomeep.imo.org. Accessed July 30, 2018. 22. Spengler, T., and G. Wilmsmeier. Sustainable Performance
and Benchmarking in Container Terminals – The Energy Dimension. In Green Ports – Inland and Seaside Sustainable Transportation Strategies (Bergquist, R., and J. Monios, eds.), Elsevier, Amsterdam, the Netherlands, 2018. 23. Ehrler, V., S. Seidel, A. Lischke, I. Y. Davydenko,
K. Dobers, A. Lewis, S. Val, and D. Luzzini. Standardiza-tion of Transport Chain Emission CalculaStandardiza-tion – Status Quo and What Is Needed Next. Proc., 7th Transportation Research Arena, Vienna, Austria, 2018.
24. Dobers, K., D. Ru¨diger, and J. P. Jarmer. Guide for Green-house Gas Emissions Accounting at Logistics Sites, 2019. http://publica.fraunhofer.de/documents/N-532019.html. Accessed February 23, 2018.
25. Higgins, C. D., M. Ferguson, and P. S. Kanaroglou. Vari-eties of Logistics Centers: Developing a Standardized Typology and Hierarchy. Transportation Research Record: Journal of the Transportation Research Board, 2012. 2288: 9–18.
26. Ru¨diger, D., A. Scho¨n, and K. Dobers. Managing Green-house Gas Emissions from Warehousing and Transship-ment with EnvironTransship-mental Performance Indicators. Proc., 6th Transportation Research Arena, Warsaw, Poland, 2016. 27. Dobers, K. Report on Progress with Transhipment Centre Calculations. Presentation at the Meeting of Global Logis-tics Emission Council (GLEC), Amsterdam, the Nether-lands, June 2017.
28. Smeets, E., and R. Weterings. Environmental Indicators: Typology and Overview. Technical Report No. 25. Eur-opean Environment Agency, Copenhagen, Denmark, 1999. www.eea.europa.eu/publications/TEC25/at_download/file. Accessed July 30, 2018.
29. Green Logistics Project. Ecological Assessment of Logistics Processes and Systems. System Definition Green Logistics – Brief Documentation. 2012. http://green-logistics-network. de/images/stories/Download/systemdefinition_summary_ en.pdf. Accessed July 30, 2018.
30. Dobers, K., G. Meyer, J. Friedrichs, G. Mauer, G. J. Schmitt, U. Clausen, M. Schneider, and D. Ru¨diger. Comparability of the Environmental Effects of Logistics Ser-vices, Sector Guidance for Ecological Assessments. Proc., 5th Transportation Research Arena, Paris, France, 2014. 31. World Resources Institute, and World Business Council for
Sustainable Development. Technical Guidance for Calculat-ing Scope 3 Emissions: Supplement to the Corporate Value Chain (Scope 3) Accounting & Reporting Standard. 2013. https://ghgprotocol.org/scope-3-technical-calculation-gui-dance. Accessed July 31, 2018.
32. EcoTransIT World. https://www.ecotransit.org/index.en.html. 33. SPM Sustainable Performance Monitor. https://spm-termin
als.com/.
34. Greenrouter. http://www.greenrouter.it/. 35. TK’Blue. https://www.tkblueagency.eu/en.
36. REff Assessment Tool. Resource Efficiency at Logistics Sites. https://reff.iml.fraunhofer.de/.
37. Lean & Green. http://www.lean-green.nl.
38. McKinsey. Distraction or Disruption? Autonomous Trucks Gain Ground in US Logistics. 2018. https:// www.mckinsey.com/industries/travel-transport-and-logistics/ our-insights/distraction-or-disruption-autonomous-trucks-ga in-ground-in-us-logistics?cid=other-eml-alt-mip-mck&hlkid= ef4a5aefe33d45e4ac99295256ed57f6&hctky=1392394&hd pid=06a4ccbb-4cca-43dd. Accessed March 6, 2019.
39. Jones Lang LaSalle. Why Multi-Story Warehouses Are Coming to America. 2018. https://www.jllrealviews.com/ industries/logistics/multi-story-warehouses-coming-amer ica/. Accessed March 6, 2019.
40. Lopez, E. 5 Multistory Warehouses Pop Up amid E-commerce Boom. 2018. https://www.supplychaindive. com/news/multi-story-warehouse-markets/539564/. Accessed March 6, 2019.
The Special Task Force on Climate Change and Energy (A0020T) peer-reviewed this paper (19-01929).
