Background and aims:
Air pollution reduction is important in health being of people and environment. Applying of an effective and efficient strategy is key consideration in facing with environmental challenges of management and control of air pollution. Recently, the main effort of environmental researchers is finding low cost and effective methods to control of environmental pollutants. Pollution prevention and control account as main techniques of pollution reduction. However, the prevention role is important than pollution control. Implementation of control practices and using of effective devices and technical equipments is necessary to achieve control strategy of pollution. One of the most important steps in air pollution control is selection the appropriate control equipment. Several factors included in selection of proper and technical equipments such as; cost, designing, efficiency, and etc. Choosing of proper control devices with lower emission rate of pollutants would result in lower destruction and environmental costs. Desired decision should be based on different criteria of socio-economic, and management. Choosing of appropriate control devices by consideration of various characteristics is complicated. Recently, one of the management challenges is selection of control devices based on the socio-economical, technical and environmental criteria. MADM models are one of the supportive decision systems applied for priority, and selection of best alternative. These methods seek for choosing the best and effective alternative by considering of multiple criteria. The aim of the present study is selection of air pollution control technologies with considering appropriate aspects of socio-economic and techno-practical criteria. Selection of suitable and effective control devices is key factor in prevention of losing human and financial capital.
Methods:
In this research, MADM technique based on fuzzy TOPSIS method has been used for appropriate selection of air pollution control equipment and pollution management. By literature review and cooperation of air pollution experts, common technologies in fields of air pollution included in petrochemical industries have been identified and keyed. Firstly, primarily questionnaire is prepared and then its reliability and validity were done using Lawshe's method. Five criteria including filtration efficiency, cost, maintainability, design ability and size for rating the control technologies related to air pollutants include of (NOX, SOX and CO) have been considered. Criteria weighting has been done using Shannon entropy and Fuzzy TOPSIS technique has been applied for prioritize of indicators. Data bank of primarily questions including pollution control technologies about three categories of air pollutants (NOx, SOx, and CO) has been collected and organized to provide final questionnaire. Then, questionnaire has been distributed between expert panel with air pollution discipline to complete it. After that, the weight of study criteria (including; cost, treatment efficiency, size, maintenance, design and build up) is identified using Shannon entropy method. The answers of expert panel to questionnaire were verbal. So, there was a need to be converted as scales with analytical property. As, doing mathematical operations on descriptive qualitative variables were impossible. Therefore, descriptive variables have been converted to fuzzy scales. In this research, triangular fuzzy values have been used. Fuzzy TOPSIS technique has been applied for ranking of indicators. Accordingly, it is implemented during 7 steps including; choosing proper fuzzy scale of measurement, provide decision matrix (calculation of importance degree of each indicator for every criterion), weighting of decision matrix, normalize weighted decision matrix, determination of positive and negative fuzzy ideal responses, identify distance of each indicator, and determination of closeness coefficient of indicators to ranking indicators.
Results:
The results of study illustrate that final weighting of study's criteria based on expert panel opinions and using Shannon entropy technique is different for each pollutant (Table 1).
Table 1. Final weighting of study's criteria based on Shannon entropy technique in terms of each pollutant
| No. | Final weighting for each pollutant | Cost | Size | Maintenance | Efficiency | Designing |
| 1 | Wi ( NOX) | 0.212 | 0.063 | 0.047 | 0.378 | 0.300 |
| 2 | Wi( SOX) | 0.278 | 0.165 | 0.059 | 0.390 | 0.107 |
| 3 | Wi( CO) | 0.111 | 0.154 | 0.042 | 0.651 | 0.042 |
Fuzziness decision matrix of control technologies for three types of pollutants is given in Table 2.
Table 2. Fuzziness decision matrix of control technologies | Pollutant | Control technology | Cost | Size | Maintenance | Efficiency | Designing |
| NOx | Selective catalytic reduction (SCR) | (1.13, 1.52, 2.26) | (0.61, 0.86, 1.06) | (0.45, 0.67, 0.89) | (0.68, 0.88, 0. 99) | (0.4, 0. 61, 0. 82) |
| Non Selective catalytic reduction (NSCR) | (1.14, 1.48, 2.18) | (0.51, 0.75, 0.97) | (0.46, 0.73, 0.96) | (0.54, 0.74, 0.88) | (0.6, 0.83, 1) |
| Non- thermal plasma | (1.26, 1.64, 2.35) | (0.56, 0.78, 0.97) | (0.5, 0.7, 0.9) | (0.69, 0.88, 1) | (0.53, 0.73, 0.9) |
| Plasma- catalytic hybrid process | (1.12, 1.5, 2.14) | (0.55, 0.79, 1) | (0.51, 0.76, 1) | (0.73, 0.91, 1) | (0.55, 0.78, 0.96) |
| Scrubber | (1, 1.25, 1.76) | (0.53, 0.76, 0.99) | (0.51, 0.76, 0.98) | (0.49, 0.69, 0.86) | (0.6, 0.82, 0.98) |
| SOX | Scrubber | (4.67, 6.53, 8.2) | (5, 6.93, 8.6) | (4.13, 5.93, 7.47) | (6.07, 7.8, 8.93) | (5.67, 7.4, 8.6) |
| Catalytic | (2.53, 4.13, 5.93) | (4.33, 6.13, 7.87) | (3.93, 5.73, 7.27) | (5.93, 7.67, 8.73) | (4.33, 6.13, 7.53) |
| Flare | (2.87, 4.4, 6.13) | (3.4, 5.2, 6.93) | (3.27, 4.93, 6.6) | (4.6, 6.47, 8) | (4.4, 6.27, 7.73) |
| High Stack | (3.8, 5.2, 6.53) | (3.27, 4.73, 6.27) | (4.2, 5.8, 7.13) | (3.07, 4.57, 6.14) | (4, 5.53, 6.87) |
| Active carbon | (2.6, 4.2, 6) | (4.13, 5.93, 7.6) | (3.27, 4/.93, 6.73) | (4.47, 6.07, 7.6) | (4.4, 6.13, 7.67) |
| CO | Non- thermal plasma | (2.73, 4.27, 5.93) | (3.87, 5.8, 7.53) | (4.27, 6, 7.67) | (7, 8.73, 9.73) | (4.87, 6.8, 8.4) |
| Plasma- catalytic hybrid process | (3.2, 4.87, 6.6) | (4.33, 6.33, 8.07) | (4.27, 6.13, 7.87) | (6.87, 8.53, 9.6) | (5.07, 6.8, 8.2) |
| Flare | (3.33, 4.87, 6.47) | (3.13, 4.87, 6.73) | (3.4, 5.07, 6.73) | (4.33, 6.07, 7.6) | (4.13, 5.73, 7.13) |
| High Stack | (3.07, 4.6, 6.2) | (2.93, 4.6, 6.47) | (4.2, 5.93, 7.53) | (2.93, 4.4, 6.07) | (4.4, 6, 7.4) |
| Scrubber | (4.07, 5.87, 7.47) | (4.4, 6.27, 7.8) | (3.8, 5.47, 7.07) | (5.53, 7.07, 8.13) | (5.07, 6.67, 8) |
After weighting of decision matrix it has been normalized and then positive and negative fuzzy ideal answers are determined. Then, distance of each indicator has been identified and closeness coefficient of indicators is reported to ranking of control technologies based on pollutant type (Table 3).
Table 3. Priority of selected optimum technologies for air pollution treatment in present study | Control technology of NOx | +D | -D | CCi | Rank | Control technology of SOx | +D | -D | CCi | Rank | Control technology of CO | +D | -D | CCi | Rank |
| Selective catalytic reduction (SCR) | 0.104 | 0.143 | 0.5782 | 4 | Scrubber | 0.272 | 0.201 | 0.4254 | 4 | Non- thermal plasma | 0.011 | 0.313 | 0.9657 | 1 |
| Non Selective catalytic reduction (NSCR) | 0.102 | 0.142 | 0.5808 | 3 | Catalytic | 0.038 | 0.435 | 0.9194 | 1 | Plasma- catalytic hybrid process | 0.022 | 0.302 | 0.9316 | 2 |
| Non- thermal plasma | 0.046 | 0.199 | 0.8129 | 1 | Flare | 0.172 | 0.304 | 0.6384 | 3 | Flare | 0.217 | 0.108 | 0.3327 | 4 |
| Plasma- catalytic hybrid process | 0.054 | 0.189 | 0.7769 | 2 | High Stack | 0.382 | 0.098 | 0.2049 | 5 | High Stack | 0.313 | 0.011 | 0.0341 | 5 |
| Scrubber | 0.181 | 0.063 | 0.2584 | 5 | Active carbon | 0.123 | 0.351 | 0.7407 | 2 | Scrubber | 0.138 | 0.211 | 0.6049 | 3 |
As we can see from Table 3, Cold plasma technology with closeness coefficient of 0.8129 was identified as the most appropriate filtration technology for treatment of NOX pollutant at the petrochemical industry. It has been choose as proper and selective option due to selective and effective treatment efficiency of this technology for NOx pollutant, very low energy consumption and safe performance of reactor because of very low amperage. Plasma- catalytic hybrid process has been selected as secondary priority because of costly selective catalyst and also generation of unwanted byproducts. Catalytic based methods with closeness coefficient of 0.9194 possess higher ranking for SOX control compared to other techniques and carbon active technology placed at the secondary rank. The main reasons in catalytic priority are high efficiency, availability of absorbent chemical materials, easy and low cost of designing and build up that technology. In priority of control technologies for CO removal, non- thermal plasma technology with closeness coefficient of 0.9679 has been identified as the best technology and plasma catalytic hybrid process placed at the next priority.
Conclusion:
It has been known that one dimensional attitude to proper decision-making process is inefficient. Also, comprehensive considerations in management methods to make appropriate decisions are undeniable. Fuzzy TOPSIS technique is a reliable method to achieve effective and efficient technology based on different and multiple criteria. Selection of appropriate control equipment using multi attribute decision making methods and consideration of multiple criteria using Fuzzy TOPSIS method are possible. In present study, selection of appropriate control technologies using Fuzzy TOPSIS technique is addressed for some pollutants including NOX, SOX and CO. Recognition, prioritize, and selection of proper technologies with low cost and effective performance in field of environmental pollution control according to effective criteria are undeniable requirements of optimum consumption of project's resources. Thus, new designs of control technologies with emphasize on effective efficiency should be identified. Multi attribute decision making method known as efficient tool in this subject. It is applied for prioritize treatment technologies with low uncertainty especially, in petrochemical industries with high level of some considerations including; socio-economic, techno-operational, and environmental.
