The evaluation of the sustainable transport system development with the scenario analyses procedure

1. Introduction

The development of sustainable transport systems is a frequently discussed subject in literature and recent scientific research. A sustainable transport system should be reliable, efficient, safety and at the same time should have a minimal environmental impact [1]. Unfortunately, due to the complexity of the transport system, experimenting on real systems is very difficult and very risky. Any organizational, technical or infrastructural changes should be preceded by a thorough analysis of the effects that will have on the functioning of the system, i.e. the economic, social and environmental effects. Limited availability and the ability to use analytical methods in the process of transport system development strategies impose the need to develop simplified models for particular transport modes. However, in order to accurately reflect the functioning of the real transport system, models should include the specific characteristics of each transport system and, above all, the interactions between them. Therefore, it is necessary to develop simulation models that allow for a comprehensive evaluation of the solutions that may affect the functioning of such systems.

The aim of the article is to present the possibility of using scenario analysis in simulation models and to show the impact of implemented changes on the functioning of the transport system and its impact on the environment. Researchers exploring these types of systems create models at different detail levels to fit the research objectives [2-5]. The development of transport systems, regardless of scale, requires tools to assist in modelling and taking into account different criteria. One such tool is the PTV VISUM environment that allows for the creation of dedicated models supporting the development of transport systems in the multiobjective and multivariate aspects. In this article, the research was based on the results obtained in the EMITRANSYS model. It is a model, prepared and tested as an efficient support for multiobjective analyses of sustainable transport systems development through scenario analysis.

Well-designed models can give a precise answer to the transport system's performance and impact on the environment and thus be used to shape transport policy. The impact of transport on the environment is an important criterion in planning activities aimed at improving its quality. The negative impact on the environment and the same on external costs of transport activities are dependent on transport volumes, a condition of infrastructure and technical equipment and also on a distribution of traffic into modes of transport within given network and expeditious switching it between the modes [6]. Evaluation of transport system development scenarios is a complex task and should be solved at the level of greater detail, including: required time, resources, regulations forcing or allowing switching modes, according to circumstances [2, 7-9].

Infrastructure investment has a major impact on improving the quality of transport systems. However, such actions are expensive and require spatial resources. This is especially important in urban systems where space is limited, and investments are very expensive. Therefore, it is necessary to take action on the proper organization of the traffic, to ensure the competitiveness of transport systems and to plan the nodal elements for goods and passenger transition between modes and places to enter and leave the system. This approach is characteristic for modern transport systems planning, which provides a harmonious interaction between different modes of transport [1, 7, 10-13].

Evaluation of the quality of the transport system requires defined criteria. These criteria cover the complexity of the transport system in an economic, social and environmental context. Each transport policy, whether a country or a region, determines the direction in which the system should be developed. Under the transport policy and the opportunities offered by the transport system, there are transport service providers whose offer is closely related to the quality of the transport system [2, 7, 14-15].

As a result, the growing demands of transport system users has led to the need to improve their operations and thus to build modern models that incorporate dynamic factors, including just in time strategies, increasing need of mobility and environmental constraints. Therefore, it is necessary to take into account the interests of all participants in the transport process. There are necessary significant improvements in terms of activities to reduce the negative impact of transport on the environment, i.e. harmful substances and noise emissions and their inclusion in external costs [1, 2, 15-17].

The article presents an example of an assessment of the urban transport system prepared on the base of MTAW 2015 model developed for Warsaw Traffic Research in 2015 in PTV VISUM. The experience gained through the development of the EMITRANSYS model has been used.

2. Scenario analysis

3. Procedure for scenario analysis

The procedure for scenario analysis is divided into three stages according to the systematic analysis (Fig. 1):

I – formulation of research objectives (technical, economic, social, environmental),

II – building a model (with appropriate tool or techniques) supporting scenario analyses,

III – evaluation of scenarios based on a multi-criterion evaluation of modelled situation.

The problem formulated for scenario analyses must take into account not only the set of goals to be achieved defined by evaluation criteria, but also the purpose of analyses at all. Long-term and short-term analysis of different types have different purposes resulting from technical, economic, organizational and legal determinants that should be taken into account when defining an appropriate model for the development of transport system (stage II of scenario analysis).

Especially the stage II (Fig. 1) is important and complex. Model must contain features of transport system common for all scenarios, be scalable and ensure a reasonable amount of certainty regarding the change in system elements.

4. Integrated index evaluation for the scenario development

An important step in the procedure is establishing a system of criteria in relation to decision goals. Criteria represent interests of decision makers and system users. The set of criteria should reflect key aspects of undertaken decisions and be universal for all scenarios. Assessing effects (positive or negative) of scenarios, with regard to uncertainty factors, boils down to formal comparative analysis of different scenarios of development according to the values of defined criteria.

Different methods can be applied to select best development option, but multicriteria methods are preferred as long as they provide objective evaluation and possibility of matching seemingly incomparable criteria. Multicriteria methods led to aggregated criterion used for the evaluation of the variant of development scenario of the transport system.

To define this criterion, set of development scenarios of transport system is defined as $\mathbf{S}=$ {$s:s=$ 1, 2, …, $N$}. Each scenario includes different variants of solution (Fig. 1) given by set $\mathbf{V}\left(s\right)=\left\{v_1\right(s)$, $v_2\left(s\right)$, …, $v_N\left(s\right)$}.

Scenarios are described by goals and criteria measuring those goals. Set of goals is defined as $\mathbf{G}\mathbf{S}=$ {$gs:gs=$ 1, 2, …, $GS$}. Each goal is described by weight $q\left(gs\right)\in$ (0, 1); weights sum up to 1. Goals are defined by sets of criteria measuring them $\mathbf{F}\left(gs\right)=$ {1, 2, …, $f\left(gs\right)$, …, $F\left(gs\right)$}. Each criteria is described by weight $q\left(gs,f\left(gs\right)\right)\in$(0, 1); weights sum up to 1 for each goal $gs$.

Then, weight of particular criteria for whole evaluation system for given scenario is expressed as $q\left(gs\right)\cdot q\left(gs,f\left(gs\right)\right)$. Criteria are normalized to allow their comparison. $n\left(v\left(s\right),gs,f\left(gs\right)\right)$ is then a normalized assessment of $s$th scenario according to $f\left(gs\right)$th criterion in $gs$th goal. On that base aggregated evaluation index of $v\left(s\right)$ development scenario for transport system can be defined as follows:

And used for evaluation purposes.

Fig. 1Algorithm for scenario research Source: [14]

5. Application of scenario analysis for evaluation of proecological solutions in the aspect of infrastructure investments – case study

5.2. Results for defined development scenarios

Traffic distribution for 2015 in MTAW 2015 is shown in Fig. 2.

The level of emission of harmful compounds for the first scenario (2015) is shown in Fig. 3. Total emission is presented in Table 2.

Noise level for 2015 is shown in Fig. 4.

Fig. 2Traffic distribution for 2015. Source: MTAW 2015 in PTV VISUM

Fig. 3Emission of pollutants for the traffic distribution in first scenario. Source: MTAW 2015 in PTV VISUM

Table 2Emissions of pollutants for the traffic distribution in first scenario

Harmful compound	Cars	LGV	HGV
CO2 [g]	311 595 087	25 295 480	53 174 954
CO [g]	567 254	29 788	113 762
HC [g]	43 039	4 533	7 884
NOx [g]	621 714	105 510	318 413
PM [g]	10 925	4 090	5 822
Source: MTAW 2015 in PTV VISUM

Fig. 4Noise level for the traffic distribution in first scenario. Source: MTAW 2015 in PTV VISUM

Fig. 5Traffic distribution for 2016. Source: MTAW 2015 in PTV VISUM

Traffic distribution for 2016 in MTAW 2015 is shown in Fig. 5.

The level of emission of harmful compounds for the second scenario (2016) is shown in Fig. 6. Total emission is presented in Table 3.

Noise level for 2016 is shown in Fig. 7.

Table 3Emissions of pollutants for the traffic distribution in second scenario

Harmful compound	Cars	LGV	HGV
CO2 [g]	321 754 869	25 329 559	53 600 135
CO [g]	540 066	26 265	105 318
HC [g]	40 124	4 172	6 543
NOx [g]	600 366	101 103	263 722
PM [g]	10 113	5 145	3 272
Source: MTAW 2015 in PTV VISUM.

Fig. 6Emission of pollutants for the traffic distribution in second scenario. Source: MTAW 2015 in PTV VISUM

Fig. 7Noise level for the traffic distribution in second scenario. Source: MTAW 2015 in PTV VISUM

Fig. 8Traffic distribution for 2020 Source: MTAW 2015 in PTV VISUM

Traffic distribution for 2020 in MTAW 2015 is shown in Fig. 8.

The level of emission of harmful compounds for the third scenario (2020) is shown in Fig. 9. Total emission is presented in Table 4.

Noise level for 2020 is shown in Fig. 10.

Obtained results show that infrastructure development in 2016 has contributed to reduced negative environmental impact of transport in the city in comparison to the first scenario as a reference point. This does not apply to carbon dioxide emissions for passenger cars and goods vehicles, as well as particulate matter for LGV. Infrastructure development in 2020 seems to be beneficial because the impact on the environment is considerably lower than in other options. The largest decrease was noted for HGV sector.

Table 4Emissions of pollutants for the traffic distribution in third scenario

Harmful compound	Cars	LGV	HGV
CO2 [g]	2 85 394 853	21 181 387	52 305 337
CO [g]	404 999	15 724	85 782
HC [g]	28 769	2 679	3 922
NOx [g]	439 007	67 698	136 668
PM [g]	5753,075	2699,509	1504,012
Source: MTAW 2015 in PTV VISUM.

Fig. 9Emission of pollutants for the traffic distribution in third scenario. Source: MTAW 2015 in PTV VISUM

Fig. 10Noise level for the traffic distribution in third scenario. Source: MTAW 2015 in PTV VISUM

A detailed juxtaposition of values for 2016 compared to 2015 is presented in Table 5, while the values for 2020 compared to 2015 in Table 6. It should be noted that the peaks in emissions on particular sections result, among others, from traffic congestion (e.g. due to the modernization of two lanes on the three-lane road).

Percentage share of active transportation network links in the whole number of active links with appropriate noise level is shown in Table 7.

Obtained results show that infrastructure development in 2016 has contributed to reduced share of active links in transport network with noise level more than 50 dBA. The equivalent level of noise significantly decreased. Comparing 2015 and 2020, it should also be noted that the number of active links with a noise level greater than 50 dBA is also reduced.

Table 5Harmful compounds emission level in 2016 compared to 2015

Harmful compound	Cars	LGV	HGV
CO2 [g]	103 %	100 %	101 %
CO [g]	95 %	88 %	93 %
HC [g]	93 %	92 %	83 %
NOx [g]	97 %	96 %	83 %
PM [g]	93 %	126 %	56 %
Source: MTAW 2015 in PTV VISUM.

Table 6Harmful compounds emission level in 2020 compared to 2015

Harmful compound	Cars	LGV	HGV
CO2 [g]	92 %	84 %	98 %
CO [g]	71 %	53 %	75 %
HC [g]	67 %	59 %	50 %
NOx [g]	71 %	64 %	43 %
PM [g]	53 %	66 %	26 %
Source: MTAW 2015 in PTV VISUM.

Table 7Percentage share of transportation network links

Noise level	2015	2016	2020
≤ 25 dBA	0,2 %	0,3 %	0,1 %
≤ 50 dBA	26,4 %	37,6 %	34,7 %
≤ 75 dBA	70,8 %	59,8 %	63,3 %
> 75 dBA	2,7 %	2,4 %	1,9 %
Source: MTAW 2015 in PTV VISUM.

6. Conclusions

The research presented in this article shows, that scenario analysis enables carrying out multifaceted analyses of cargo and passengers flow in the transport network. To perform such analysis, the simulation tools are required. One of the universal tools that can be used in this type of problem is the PTV VISUM environment. This tool allows analysis of scenarios of transport system takin into account different technological, economic and environmental criteria. Noise level considered in this article can be used as one of these criteria (EMITRANSYS, MTAW 2015). Extensive software capabilities share the possibility of detailed analysis and evaluation of alternative traffic distribution in different scenarios of development. Such analyses can then facilitate decision-making, regarding transport infrastructure investment in a given area of the transport network. Additional economic projections may indicate the direction of changes in the demand for transport services.

Because each participant in the decision-making process can represent different partial criteria functions, it is difficult to indicate a solution that will satisfy each transport system user. That is why the multi-criteria methods gain interests according to assessment of decision based on contrary criteria, which represent different expectations of transportation network users. In practice, the implementation of the multi-criteria objective function is problematic, because a mutual conformity of partial criteria of the objective function appear only in rare cases.

Thus, the inclusion into the multi-criteria mathematical programming model more than a single criterion will not lead to the unequivocally best solution from the point of view of all the partial criteria at the same time.