Ecologically preferred types of drive systems for city buses—the context of the Polish energy mix

The scope of the analysis presented in this paper is focused on profitability analysis of using different drive systems of urban buses depending on the emissivity of the energy system in a country. The government administration of each country sets its own policies to support individual propulsion sources, and thus significantly influences decisions made by transport companies. From a market point of view, it is difficult to require an entrepreneur to make decisions through the prism of environmental impacts. Every entrepreneur would rather seek to maximise profits. The state government, on the other hand, must pay attention to the environmental impact of vehicles in operation, because it is the residents who bear the environmental, and thus health, costs of vehicle operation. Consequently, it is the government that should introduce relevant support mechanisms in order to shape the demand for those vehicles which, in the opinion of the government, are the least costly, also taking into account the environmental impacts. The main objective of this study is to provide the methodology that is universal for EU countries and helpful in creating a transport policy for supporting the purchase of city buses powered by particular types of propulsion.

The analyses have been carried out only in regards to the usage stage of public transport buses, without taking into account the process of their production and transport or their disposal. This is due to the undertaken approach, which leans towards analysing the environmental impact from a country-specific point of view. Given that the production of a bus often takes place in a different country than that of the bus usage and disposal (e.g., for urban buses there exists the secondary market), the only stage in the life cycle of an urban bus which economic cost can be directly compared in each country is the vehicle usage (operation) stage. As a result, the proposed methodology applies only to this stage of the vehicle’s life cycle. If a given country is the manufacturer of a particular bus type and/or takes into account the disposal of this means of transport, it is necessary to carry out additional environmental cost analyses for these two life cycle stages so as to obtain the total environmental costs for a particular vehicle type.

The proposed methodology is applicable in EU countries, as the environmental cost analyses assume calculations on the basis of pollutant emissions, compliant with the EURO standards in force in the EU. Under market conditions, which are not limited by EURO emission standards, the environmental costs of operating new vehicles may be different, due to possible different emission limit values for individual pollutants.

All calculations were made for 2021, as the latest available data, at the moment of the publication, containing all the indicators used in the analyses from 2021 within the reporting under the Emissions Trading Scheme [KOBiZE 2022].

The presented framing of this study is relatively new and adds to the emerging body of the literature on environmental and economic implications of the types of urban bus drive systems. The literature review identified studies on case-related modelling of emissions, such as modelling the fuel consumption and pollutant emissions of the urban bus fleet of the city of Madrid [López-Martínez et al. 2017], or in the city of Stockholm [Xylia M. et al. 2019] or cost benefit analysis of three different urban bus drive systems [Gerbec et al. 2015] concerning CNG, hydraulic hybrid buses and conventional (diesel) buses. Also, Muñoz et al. [Muñoz et al. 2022] presented a comparative analysis of the cost, emissions, and fuel consumption of diesel, natural gas, electric, and hydrogen urban buses, showcasing an integrated index composed of three indices that measure well-to-wheel energy use, global warming potential in terms of carbon dioxide equivalent emissions, and total cost of ownership, applied to the case of Argentina. At the same time, Correa et al. [Correa et al. 2019] developed a comparative energy and environmental analysis of a diesel, hybrid, hydrogen, and electric urban bus within the well-to-wheel scope in Argentina, Chile, and Brazil. Our study focuses on external costs due to individual pollutants (such as: SO₂, NO_x, PM_2.5, PM₁₀) for different types of urban buses (like diesel, hybrid, plug-in hybrid, electric, CNG, and hydrogen) operating in Poland. The analysis performed is of much importance as, e.g., particulate matter (PM) is one of the pollutants most harmful to human health in Europe [Timmers, Achten 2016]. Also, according to OECD, the welfare costs of premature deaths due to PM exposure amounts to approximately 4.15% of global GDP in 2017 [OECD 2020], [OECD 2019].

In some studies, analyses relevant to urban transport policies were carried out. For instance, Dadashev et al. [Dadashev et al. 2023] performed an analysis of the effects of carbon-related transportation policies, through including changes in transport demand, fuel consumption, various emissions levels, and environmental equity impacts, into simulations of the Tel-Aviv metropolitan area transport policies. Gustafsson et al. [Gustafsson et al. 2018] investigated different ways of assessing and presenting the energy performance of public bus transport systems and assessing renewable and fossil fuels energy efficiency. Wang et al. [Wang et al. 2022] developed a modelling approach to plan low-carbon energy-transportation systems at the metropolitan scale in Beijing, China. Overall, the relevant literature is focused on environmental effects of transport policies in case studies of the cities or countries. In this article we present the developed methodology for the analysis of environmentally preferable types of urban bus drive systems in relation to the energy mix of a given country validated on the example of Poland.

The proposed methodology consists of the following steps presented in Figure 1 and described in further detail below.

Figure 1.

Step 1: Selection of specific bus types for comparison.

The selection of vehicles for comparison should reflect the popularity of the analysed bus types in a given market.

In addition, bus types should be as similar as possible in terms of functionality. It is proposed to include the following categories to differentiate between the various types of buses: –

power system

In relation to the power system, the buses selected for comparison should be equipped with different power systems, the most common ones being the following: –

diesel,

It is important that the vehicles being compared are as similar as possible. For this reason, it is advisable to compare vehicles from the same manufacturer. When it comes to size class, the most common vehicle size classes on the market are: –

MIDI (most often vehicles with a length of approximately 9 metres),

In addition to the buses from the above classes, there are other ones, both smaller—mini classes, or buses designed for passenger transport—and larger—double articulated vehicles exceeding 24 metres in length. However, taking into account the criterion of market popularity in EU countries, the application of the methodology can be limited to selecting vehicles from one of the three groups listed above.

With regards to additional design elements and equipment, the main distinguishing features of the vehicles are the level of floor routing and the type of gearbox used.

Urban buses most often come in low-floor or medium-floor versions. High-floor buses and double-decker buses may also be used, but these are designed for intercity services (high-floor) or tourist services (high-decker). Currently, high-deck buses do not appear as new, modern designs in urban traffic, nonetheless, such vehicles are still in operation in many of Poland’s cities. Given that the methodology is intended to identify support mechanisms for new buses, the choice of bus versions related to floor height should be limited in the case of urban buses to low-floor and low-capacity buses, and in the case of suburban routes, the choice may possibly be extended to include medium-floor buses.

Under the proposed methodology, an additional distinguishing feature of a given bus is the type of gearbox (manual or automatic). Yet, it is recommended to choose an automatic gearbox, as this is the predominant solution in modern urban vehicles and, besides, hybrid, electric, and hydrogen vehicles only come with this type of gearbox.

Market practice indicates that it is extremely difficult to find analogous vehicle types in all the categories analysed, as some manufacturers do not even offer analogous types with all possible propulsion sources. It is therefore likely that the most popular vehicle types on a given market will not have their equivalents in the categories analysed. In such a situation, it is possible to include different vehicle types in the analyses, and then special attention should be paid to ensure that bus practical utility is as similar as possible, for instance: bus capacity, passenger comfort (e.g., air-conditioning) or additional equipment (e.g., type of ticket dispensers, ticketing machines, and passenger information systems).

Step 2: Obtaining fuel and energy consumption data for selected bus types.

In order to define national level policy, it is desirable to obtain comparable results, so it is recommended to determine volumes on the basis of test data or the standardised on-road test (SORT), the principles of which have been developed by the World Association of Public Transport (UITP). Three types of tests have been developed: SORT 1, SORT 2, and SORT 3. The SORT 1 test corresponds to operations in a highly congested city, for SORT 2 the parameters are chosen to reflect standard urban traffic, and the SORT 3 test is designed for suburban traffic. Depending on the prevailing operating conditions, an appropriate type of test should be selected. In the absence of specific operating conditions, the SORT 2 test is recommended. In market practice, however, it is extremely difficult to obtain fuel consumption results according to the SORT 2 test for identical vehicles, which differ only in their propulsion system. This is for two reasons. Firstly, manufacturers of city buses do not publish official fuel consumption data, as this depends very much on the specific route and operating conditions of the bus. A prospective city bus buyer is most likely to test the vehicle under consideration of the routes on which it is to be used. The second reason for the lack of comparable data is the fact that the SORT tests have to be performed for a specific combination of engine, transmission, and additional vehicle equipment. In practice, this means that the SORT 2 test is performed for a specific tender for which the manufacturer is preparing, and the vehicle is tested in the configuration specified by the customer. Given these limitations and difficulties, it is possible to carry out all emissions calculations without fuel consumption data, but only for combustion vehicles. In such a situation, the option to calculate CO₂ emissions should be selected in COPERT and it will be determined for the average fuel consumption of the buses in the category. It should be noted that for battery buses recharged from the electricity grid as well as hydrogen buses, it is always necessary to have electricity consumption data, as the emissions from such buses will depend on these data and the energy mix adopted in the country under the analysis.

Step 3: Calculation of pollutant emissions (NMVOC, NO_x, PM_2.5, PM₁₀, CO₂) for diesel and/or hybrid bus types.

The above emissions for combustion and hybrid buses, in addition to CO₂ emissions, are recommended to be calculated in COPERT V. If newer versions of COPERT become available in the future, it is recommended that the latest version is used. Nationally representative values should be taken as input to COPERT, in particular for operating speeds. For the calculation of emissions, it is necessary to enter basic data on the vehicles under analysis and their operating conditions. Among the most important are the share of mileage performed under specific conditions (urban peak, urban off-peak, rural, highway) and the average speeds within a given type of infrastructure.

CO₂ emissions are proposed to be calculated on the basis of CO₂ emission factors and calorific values of the fuels analysed, using the fuel consumption of a specific vehicle type. The CO₂ emission values for internal combustion and hybrid vehicles can also be calculated with the use of COPERT, However, it should be noted that not all vehicle types are entered into the COPERT software. If a vehicle is analysed that does not have its category assigned in COPERT software (e.g., hybrid articulated bus or CNG articulated bus), it will be necessary to have it entered into another and most similar category. Therefore, the calculations may be less accurate. If the actual fuel consumption for the bus types analysed is available, it is recommended to use conversion by emission factors.

Step 4: Determination of the CO₂ emission factor from electricity generation in the country.

The CO₂ emission factor per unit of electricity produced depends on the energy mix used in a given country in a given year. These values should be available from national statistics.

Given that the support mechanism designed based on the proposed methodology will be introduced in the future and will apply to buses in operation for several or more years after their introduction, changes in the CO₂ emission factor can be projected. To this end, it is proposed to use available historical data and one of the basic statistical methods, the exponential smoothing method. It is possible to use another forecasting method if there is a factual basis and appropriate data are available.

Step 5: Calculation of CO₂ emissions from electric and plug-in hybrid buses.

The calculation of CO₂ emissions related to electricity consumption by electric buses and plug-in hybrid buses with the use of the CO₂ emission factor from electricity generation.

Step 6: Calculation of pollutant emissions (SO₂, NO_x, PM_2.5, PM₁₀) for electric and plug-in hybrid buses.

The emissions of SO₂, NO_x, PM_2.5, and PM₁₀ as a result of electricity consumption can be calculated for electric buses with the use of the emission factors for individual pollutants per unit of electricity generation (based on national data).

Step 7: Calculation of emissions related to the production of hydrogen for powering a hydrogen bus.

It was assumed that the most common way of producing hydrogen to power road vehicles is electrolysis, and thus, it is proposed to take the emissions related to hydrogen bus powering as those related to electricity generation needed to produce a certain amount of hydrogen. It is because electricity generation affects, through a specific energy mix, the emissivity of this fuel. Therefore, in this analysis, emissions related to the production of hydrogen are based on the emissivity of the energy mix. Only in cases where the electrolyzer is powered solely by energy from a purpose-built RES installation should the emissions from this process be assumed to be zero. However, in the situations when the electrolyzer is powered by energy coming solely from a RES installation that already existed and supplied energy to the electricity grid, it seems reasonable to assume that energy to power the electrolyzer comes from the electricity grid. So, the assumption in this case would be to use the emissivity of the energy mix and not the emissivity of the RES. It is because if this particular electrolyzer was not operating, the energy from this RES source would power other equipment, entering the electricity system, and thus lowering the emissivity of the overall energy mix. The aggregated overall emissivity at a level of the whole country is therefore the same, regardless of whether this particular RES installation only powers the electrolyzer or goes into the electricity grid.

In exceptional situations, when hydrogen is obtained by other means, the emission factors for the production of 1 kg of hydrogen should be determined for the technology used.

Step 8: For electric and hydrogen buses, with the use of COPERT V (or later version if available), the calculation of the emissions of PM_2.5 and PM₁₀ (“non-exhaust”), from wear and tear of brake linings, tyres, and road surface and adding the obtained values to the emissions calculated for these vehicle types in the previous steps.

For electric and hydrogen buses, PM_2.5 and PM₁₀ emissions resulting from abrasion of brake systems, tyres, and road surface should be calculated and added to the emissions resulting from generation of electricity to power the vehicle.

Emissions from brake lining abrasion were assumed to be 1/3 of the emissions for buses powered by other types of propulsion (for all other buses, COPERT assumes the same PM emissions from braking systems). This assumption is due to the fact that in vehicles with electric traction motors, the braking system is significantly less stressed than in combustion vehicles, as braking is primarily done through the powertrain as part of energy recuperation. The braking power and the amount of energy recuperated in this system depends, inter alia, on the capacity of the batteries and the level of energy stored in them, which explains the difference between purely electric and hybrid vehicles (which also have batteries, but with significantly less capacity). In reality, therefore, brake lining wear is much less than that for combustion vehicles, but it is very difficult to determine exactly how much less. Neither electric nor internal combustion buses were included in COPERT software, so it is not possible to determine the emission factors based on this methodology. Analyses carried out by different research teams around the world have assumed different reductions in brake lining wear, ranging from an assumption of no brake lining wear (100% wear reduction as compared to a conventional vehicle), through studies assuming reductions of up to 50%, to reductions as low as 25%. Under this study, there was an assumed 2/3 reduction in brake emissions, based on comparative studies of the most popular combustion and electric cars (Tesla, BMW i3, and Nissan Leaf).

PM_2.5 and PM₁₀ emissions for tyre and road surface abrasions for electric and hydrogen buses were assumed to be analogous to those for combustion and hybrid buses. In this regard, there have been studies showing that tyre and road surface abrasion emissions are higher for electric vehicles than those for combustion vehicles, which may be due to the additional mass of traction batteries. Nonetheless, due to methodological inconsistencies in these studies, it was decided to assume that there was no difference in tyre and road surface abrasion emissions.

Step 9: Calculation of the external costs associated with the operation of individual public transport bus types based on estimated pollutant emissions.

The monetisation of the calculated pollutant emissions is recommended to be carried out on the basis of the methodology “Handbook on the external cost of transport” - version 2019 (or newer if available), which is published on the website of the European Commission [European commission 2019].

Step 10: A sensitivity analysis of the environmental cost depending on the CO₂ emission factor for a given energy mix.

It is proposed to carry out a sensitivity analysis of the environmental cost for the compared vehicles depending on the CO₂ emission factor and to establish the limits of this factor at which the environmental cost-effectiveness changes to different bus types. It is proposed to check whether the limit value established in the sensitivity analysis falls within the range of the projected CO₂ emission factors in the future and, if necessary, to include this fact in the decision-making process.

In the following section, the implementation of the proposed methodology to the Polish Case Study is presented.

The energy mix of Poland in 2021 conditioning the results of the analysis is as follows, based on [KOBIZE 2022, Zalewska 2022, p.13]: –

Lignite – 25.6% (46 TWh).

The analysis followed the steps of the methodology outlined in the section above, with the assumptions described below.

Step 1: Selection of specific bus types for comparison. The following bus types were selected for the analysis: –

Urbino 12 diesel,

All the selected vehicles are produced by Solaris Bus & Coach sp. z o.o. In Poland, these are among the most frequently purchased new public transport buses, (market share in 2021—63.7%). The buses selected for the analysis are the most common class (MEGA class with a length of approximately 12 metres) and have a low floor run at all doors. They are equipped with automatic transmissions, air conditioning, and all types of propulsion are represented by the bus types operated in Poland.

Step 2: Obtaining fuel and energy consumption data for selected bus types.

Data for the SORT 2 cycle was not available for all analysed vehicles. It is also not possible to collect fuel and/or electricity consumption data for all the analysed bus types from one public transport operator, as no public transport company in Poland operates all the analysed bus types. Fuel or electricity consumption was therefore assumed on the basis of operating data from various Polish cities and on the basis of the Environmental Product Declaration (EPD), at the following level: –

Urbino 12 diesel 39.1 l/100 km [Analiza kosztów i korzyści… 2018, p.17],

Step 3: Calculation of pollutant emissions (NMVOC, NO_x, PM_2.5, PM₁₀, CO₂) for diesel and/or hybrid bus types.

Calculations of NMVOC, NO_x, PM_2.5, and PM₁₀ emissions for internal combustion and hybrid buses were carried out with the use of COPERT V software ver. 5.5. The calculations were performed for the following input data: –

Vehicle segment: Urban Buses Standard 15–18 t (for all propulsion types).

CO₂ emissions for the vehicles analysed were calculated on the basis of CO₂ emission factors and fuel calorific values [KOBIZE 2022], using the fuel consumption of the examined vehicle given in section 2.

Step 4: Determination of the CO₂ emission factor from electricity generation in the country.

The CO₂ emission factor per 1 kWh of electricity, depending on Poland’s energy mix and for the period 2014–2021, used in the calculations is presented in Table 1.

The analysis utilises the latest available data. The information is published annually in December and relates to the previous year, therefore, the data for 2022 can be expected in December 2023.

The CO₂ emission factor was projected to 2030, based on an exponential smoothing method. The projected values, for a 95% confidence interval, are presented in Table 2.

Step 5: Calculation of CO₂ emissions for electric and plug-in hybrid buses.

Using the CO₂ emission factor for 2021 (Table 1) and the electricity consumption data (from Step 2) of the electric buses and plug-in hybrid, the CO₂ emissions for these types of buses were calculated.

Step 6: Calculation of pollutant emissions (SO₂, NO_x, PM_2.5, and PM₁₀) for electric and plug-in hybrid buses.

Based on the 2021 emission factors shown in Table 3, NO_x, SO₂, and TSP emissions were calculated for electric and plug-in hybrid buses.

Step 7: Calculation of the emissions associated with the production of hydrogen to power a hydrogen bus.

It was assumed that hydrogen to power public transport buses was produced by electrolysis. This process requires 50 kWh of electricity to produce 1 kg of hydrogen. With this assumption, the emission factors from electricity production in 2021 in Poland were applied (Table 1 and Table 3) and emissions were calculated for this type of bus.

Step 8: For electric and hydrogen buses, with the use of COPERT V (or later version if available), the calculation of the emissions of PM_2.5 and PM₁₀ (“non-exhaust”), from wear and tear of brake linings, tyres, and road surface and adding the obtained values to the emissions calculated for these vehicle types in the previous steps.

For electric and hydrogen vehicles, particulate matter emissions from the wear of tyres on the road surface and brake linings were determined with the use of COPERT V software ver.5.5. These emissions were added to the emissions from the production of electricity needed to power the vehicles.

The total emission volumes for internal combustion, hybrid, electric, and hydrogen buses operating under the conditions of the Polish energy mix in 2021 are presented in Table 4.

as CO₂ emissions are several orders of magnitude higher than emis sions of other pollutants, to maintain clarity in the table, values are presented with different degrees of accuracy.

Step 9: Calculation of the external costs associated with the operation of individual public transport bus types on the basis of estimated pollutant emissions.

Based on the emissions shown in Table 4, the external costs in 2021 were calculated using the methodology described in the Handbook on the external cost of transport ver. 2019 and expressed in 2020 prices using the Harmonised Indices of Consumer Prices (HICP), as illustrated in Figure 2. Under the conditions of the Polish energy mix in 2021, the lowest external cost is associated with the operation of a hybrid bus. The second in rank is a plug-in hybrid vehicle, the third is a bus powered by ON, and the fourth is an electric bus. A significantly higher external cost is related to the operation of a CNG-fuelled bus, due to the increased cost of NO_x emissions, and the highest environmental cost—to a hydrogen vehicle, due to very high CO₂ emissions from hydrogen production.

Figure 2.

Given that the policy to support public transport is a long-term programme, the lifetime of city buses is set at 12–15 years, and given that the emission factors will vary from year to year, it seems advisable to analyse how a change in the most relevant parameters will affect the environmental cost ratio of the different bus types. In order to determine which of the analysed parameters have the greatest impact on the external costs, the shares of the external costs of individual pollutants in the total external cost were determined for all analysed vehicles, as shown in Figure 3. On the basis of the above, it can be concluded that in all cases, CO₂ is responsible for the largest costs, so it is advisable to carry out a sensitivity analysis for the CO₂ emission factor.

Figure 3.

Step 10: A sensitivity analysis of the environmental cost depending on the CO₂ emission factor for a given energy mix.

A sensitivity analysis of the environmental cost depending on the CO₂ emission factor was carried out to specify at what value of the CO₂ emission factor the environmental cost of the operation of one vehicle type would be equal to the environmental cost for another type of vehicle. The results of the analysis carried out in this way are shown in Table 5.

The obtained values of the CO₂ emission factors were then compared with the projected values of this factor presented in Table 2. If the value of the factor at which the environmental costs equalise can occur in the years covered by the projection, it means that in the time covered by the analysis, it is possible that a given type of vehicle will become more environmentally attractive than that with which it was compared. From the decision-maker’s point of view, this means that public transport support policies should be designed to allow a change of preference for the types of buses in the future, depending on the fulfilment of certain conditions at a given time.

When analysing the above data, it can be seen that the equalisation of external costs for an electric bus with an ON bus will occur with the CO₂ emission factor 0.692. Such a value, according to the forecast, should be reached in 2022. Similarly, when analysing Table 5, an electric vehicle should achieve lower external costs than a plug-in hybrid by 2028 and lower external costs than a hybrid by 2029. The other values in Table 5 do not fit into the projections (Table 2), so it can be assumed that a hydrogen vehicle will not generate lower environmental costs than hybrids or ON vehicles in the near future.

The energy mix used in a given country significantly influences the environmental attractiveness of individual propulsion types, which is why it is not possible to unambiguously identify the most environmentally friendly propulsion system for urban buses for the European Union as a whole.

Each state, when designing a public transport support policy, should be guided by their own specific technical and market parameters, as only such an approach would enable the optimisation of measures.

In the case of Poland, where the energy mix is predominantly based on coal, hybrid vehicles are the most environmentally efficient. At the same time, ON vehicles are less environmentally burdensome than electric vehicles. Hydrogen vehicles, due to the need to supply a large amount of electricity to produce hydrogen, are the least environmentally attractive.

Changes in the energy mix may result in changes in the attractiveness of the different propulsion types. According to forecasts, the electric vehicles will become more environmentally attractive in Poland as the CO₂ emission factor decreases. In 2022, electric vehicles should be better in emission savings than ON vehicles, and by 2028 and 2029, successively better than plug-in hybrid and hybrid vehicles.