Approaches on the urban traffic intensity impact on human factor determination.

Blaga, Florin Sandu ; Hule, Voichita Ionela ; Tarca, Ioan Constantin 等

Abstract: The paper proposes an evaluation method of sonic pollution and vibrations impact on human beings. The method uses fuzzy sets specific techniques. A fuzzy-type evaluation system was defined, having as inputs: traffic intensity at a specific location in Oradea city, congestion degree (population density) in that specific location.

Key words: fuzzy, traffic intensity, impact indicator.

1. INTRODUCTION

Sonic pollution is a significant characteristic of the urban conglomeration, having harmful effects on human beings. The pollution degree evaluation became a major desideratum for the local authorities. Their purpose is therefore to take firm measures regarding the decreasing of the acoustic pollution. Having this purpose in view a decisional support which should permit efficient solutions is needed.

The Romanian Government has adopted, in the Order no. 678/2006, the calculus methods for the noise indicators caused by the road, railroad, flight traffic, and also by the industrial activities, recommended by the European Union. Also, guidance lines regarding strategic noise maps creation is presented. In this framework the estimation of the number of peoples exposed to different noise levels is presented. A method to calculate the sonic pollution impact evaluation upon the residents in affected areas is not yet provided.

This paper proposes an evaluation method of the sonic pollution impact evaluation upon the human beings, using fuzzy sets specific techniques (Barron, 1993).

Fuzzy sets offers the possibility to define multi-attribute type decisional systems which can take into account a multitude of factors (criteria) having a high degree of uncertainty during evaluation (Klir & Bo, Y. , 1995). In our case these factors are: traffic intensity and population density. Furthermore, these factors can be connected through fuzzy sets specific operators, so that the decisional system output, impact indicator, should reflect the combined influences of the control inputs.

2. IMPACT EVALUATION PROCEDURE SPECIFIC TO FUZZY SYSTEM

The steps needed to be followed in order to realize the decisional process based on fuzzy sets will be described.

1. Control input in report to which the Impact Indicator (II) will be determined. Input controls in respect with which the Impact Indicator is determined are: Traffic Intensity (TI) and Population Density (PD). They form the evaluation criteria set: IN = {TI, PD}.

2. The definition of the range for each of the evaluation criteria. A range is associated to each control input, containing all the input values. These domains are:

TI: [D.sub.TI] = [L.sup.inf.sub.TI], [L.sup.sup.sub.TI]

PD: [D.sub.PD] = [L.sup.inf.sub.PD], [L.sup.sup.sub.PD] (1)

where [L.sub.inf] and [L.sup.sup] are inner and outer limits of the range associated to each control input

3. Linguistic variable associated to each control input definition. A linguistic variable is associated to each control input. To simplify notations, the linguistic variable should have the same name as the control input. So, control inputs becomes linguistic control inputs

4. Linguistic degrees associated to each linguistic variable. For each output linguistic, linguistic degrees (Sofron et al 1998) or linguistic terms are defined (Preitl. & Precup 1997). These degrees will serve to the vague characterization of the firm information. The linguistic degrees sets associated to each of the linguistic control output will have the form:

TI: [LD.sup.TI] = {[LD.sup.TI.sub.1], [LD.sup.TI.sub.2], ..., [LD.sup.TI.sub.K]}

PD: [LD.sup.PD] = {[LD.sup.PD.sub.1], [LD.sup.PD.sub.2], ..., [LD.sup.PD.sub.K]} (2)

A membership function is associated to each linguistic degree, corresponding to one linguistic control input.

5. The control output of the decisional process definition. The control output of the decisional process is the Impact Indicator (II).

6. The control output range definition. The control output range Impact Indicator (II) is:

II: [D.sub.II] = [[L.sup.inf.sub.II], [L.sup.sup.sub.II]] (3)

7. The definition of the linguistic variable corresponding to the control output. Impact Indicator (II) linguistic variable is associated to the Impact Indicator (II) control output.

8. Linguistic degrees associated to the control output values. Linguistic degrees, or linguistic terms are defined for linguistic variable associated to the control output. These degrees will vaguely characterize the information result during the inference process. Linguistic degrees set associated to the linguistic control output looks like:

II: [LD.sup.II] = {[LD.sup.II.sub.1], [LD.sup.II.sub.2], ..., [LD.sup.II.sub.K]} (4)

For each linguistic degree corresponding to a linguistic variable that describes a control output a membership function is associated.

9. Membership functions connection method. Inferential machine. Linguistic variables and linguistic degrees sets having associated membership functions, vaguely characterize the control input and the control output values.

The inference machine is realized with a set of rules as follows:

IF (premise) THEN (conclusion)

Premise--is a constant property emerged as a result of linguistic degrees connection of the control input values through procedures that are specific to the fuzzy sets theory.

Conclusion--is the stipulated property and it will be express through linguistic degrees associated to the control output linguistic variable.

10. The defuzzyfication method. Defuzzyfication is the operation of a "crisp" value achievement for the control output, using the "result" membership function of the fuzzy inference. Among the multitude of the defuzzyfication methods, the center of weight method will be used, as it is the most used method in all applications.

3. THE IMPLEMENTATION OF THE EVALUATION SYSTEM BASED ON FUZZY SETS

Evaluation system based on fuzzy sets was implemented in Toobox Fuzzy Logic from the Matlab software.

Linguistic variables associated to these two controls are Traffic Intensity (TI) and Population Density (PD). Fig. 1 shows these control inputs, linguistic degrees and membership functions associated to each linguistic degree. Therefore linguistic control Traffic Intensity (TI) has the following linguistic degrees associated: very small (vs), small (s), medium (Md), intense (I); very intense (VI).

Linguistic control Population Density (PD) has the following linguistic degrees associated: very small (vs); small (s); medium (Md); high (H); very high (VH).

The membership functions for the TI control input are triangular in shape for the small, medium and intense linguistic degrees and trapezoidal shaped for the linguistic degrees: very small and very intense. For the PD control input the membership functions are triangular shaped for the linguistic degrees small, medium and high and trapezoidal shaped for the linguistic degrees very small and very high.

Fig. 2 shows the control output Impact Indicator (II), linguistic degrees and membership functions associated to each of the linguistic degree. The linguistic degrees are: very small (vs), small (s), medium (Md), high (H), very high (VH).

The membership functions are triangular shaped for the linguistic degrees: small, medium and high and trapezoidal shaped for the linguistic degrees very small and very high.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

The inference motor is composed of 25 rules like these ones mentioned below:

1. If (IT is fm) and (DP is fm) then (II is fm) ...

12. If (IT is Md) and (DP is m) then (II is Md) ...

25. If (IT is FI) and (DP is FM) then (II is FM)

4. CASE STUDY

To evaluate the Impact Indicator ten points were identified in Oradea town where the traffic intensity was studied counting the passing auto vehicles during one day. Also population density (residents) in these locations was estimated. These data were the control input in the evaluation system and they are presented in table 1.

Using the evaluation system for these ten control input sets, Impact Indicator values corresponding to the chosen points are obtained:

II={0,9173; 0,9135; 0,8179; 0,7974; 0,7885; 0,7630; 0,6920; 0,7500; 0,500; 0,4279}.

It can be seen that in the intense traffic zones with high population density the impact indicator has greater values.

5. CONCLUSIONS

The method proposed by the authors can offer useful information to the decisional factors in identifying the urban zones with a high degree of sonic pollution risk harmful to the population.

This information should be the decisional support to establish the acoustic attenuation measures: traffic decrease level, acoustic insulation of the buildings etc.

Further developments of the method should take into account the control inputs "refinement": differentiation of the traffic type (heavy, light) population differentiation (residents, workers). They will permit an increase of the results accuracy.

6. REFERENCES

Barron, J. J. (1993), Putting Fuzzy Logic into Focus, Byte, Vol. 18, No. 4, (April,1993), ISSN:0360-5280, pages: 111-118

Klir, G. J., Bo, Y. (1995), Fuzzy Sets and Fuzzy Logic: Theory and Applications, Prentice Hall PTR, ISBN-10: 0131011715, ISBN-13: 978-0131011717

Preitl, St. & Precup, E. (1997), Introduction to fuzzy control (in Rumanian), Ed. Tehnica, ISBN: 973-31-1081-1, Bucharest

Sofron, E., Bizon, N., Ionita, S. & Raducu, R. (1998), Computer aided fuzzy control systems. Modeling and design (in Rumanian), Ed. All, Bucharest

Romanian Government (2006), Order No. 678/2006, Bucharest

Table 1. Control input in the evaluation system.

Point Street Name IT DP

1 Dacia 4236 900
2 Pod Continental 3912 800
3 Calea Aradului 3100 850
4 Decebal 2928 950
5 Nufarului 2863 900
6 Cantemir Dimitrie 2586 850
7 Pod Decebal 2378 700
8 B-Dul Stefan Cel Mare 2252 800
9 Clujului 2132 500
10 Decembrie 1664 450