Selection of Medical Waste Disposal Method for a University Hospital Using Hybrid Multi-Criteria Decision-Making Methods: A Case Study in Adana Province, Turkey

Abstract

The global expansion of healthcare services has made medical waste management an increasingly critical and complex issue. Medical wastes require specialized management due to their high infection risk, potential for environmental pollution, and adverse effects on public health. The correct collection, transportation, and final disposal are vital for protecting environmental health and ensuring the safety of hospital personnel and the community. Numerous disposal methods exist. Selecting the appropriate one, however, is a multi-dimensional decision-making problem, necessitating the simultaneous evaluation of various conflicting criteria. Adana, one of Turkey’s largest provinces, generates significant medical waste volumes due to its dense population and developed health infrastructure. Therefore, choosing the most suitable disposal method for hospitals in Adana is crucial for establishing an effective and sustainable waste management system. Making this decision using traditional methods is difficult. The multitude of criteria prevents any single method from being optimal across all aspects. This complexity mandates the use of Multi-Criteria Decision-Making (MCDM) methodologies. In this study, MCDM methods were applied, based on expert opinions, to select the disposal method at a university hospital in Adana. The research examined twelve criteria and four alternatives. The CRITIC (Criteria Importance Through Intercriteria Correlation) method was employed to objectively weigh the criteria. For the rigorous evaluation and ranking of the alternatives, three robust MCDM methods were utilized: PROMETHEE (Preference Ranking Organization Method for Enrichment Evaluation), TOPSIS (Technique for Order Preference by Similarity to Ideal Solution), and EDAS (Evaluation based on Distance from Average Solution). The final results conclusively identified incineration as the most appropriate disposal method for the hospital.

1. Introduction

Although advances in medical science and technology have brought great benefits to the protection and improvement of human health, the management of wastes from these processes has become an increasingly complex issue globally. The increase in the human population has particularly triggered a great surge in the amount and type of medical wastes from healthcare institutions. Medical wastes are wastes from hospitals with operating theaters, laboratories, patient rooms, and outpatient clinics, from medical practices and medical centers, assisted living facilities and nursing homes, and from ambulance services. These wastes are basically divided into different categories such as ‘sharps wastes’ (needles, scalpels), ‘pathological wastes’ (body fluids, organs), and ‘infectious wastes’ (blood-contaminated materials) [1]. Each category carries a specific risk profile. For instance, infectious waste is a primary risk factor for the spread of infectious diseases such as tuberculosis and hepatitis B and C, while sharps waste can accelerate the spread of these diseases through physical injuries. These risks can affect not only healthcare workers, but also waste collection and transport workers, staff working in waste disposal facilities, and even the general public due to improper disposal methods. When contaminated waste gets into the soil or water, it can damage the ecosystem, build up in the food chain, and cause serious environmental problems in the long run.

Due to the resulting infection risks, potential for environmental pollution and the foregoing negative impacts on public health, the proper collection, transport, and final disposal of medical wastes without harming the environment and human health is critical for both protecting environmental health and ensuring the safety of hospital staff and the community. It requires a special management approach. Medical wastes inflict more costs than solid wastes during disposal [2]. Therefore, the choice of disposal method is critically important in medical waste management. The selection of disposal method for medical wastes is not a simple process based on a single criterion. Rather, this decision requires the simultaneous assessment of a number of conflicting factors such as environmental impacts, economic costs, technological feasibility, regulatory compliance, social acceptability, and public health risks. For instance, the most environmentally friendly disposal method may be the costliest, or the most economical method may be the one with the greatest environmental risks. Determining the most appropriate disposal method is a critical decision-making process. In this complex decision-making environment, MCDM methods provide decision-makers with a systematic and rational framework. The objective of the MCDM methods is to identify the most appropriate option by taking into account multiple criteria and the relationships between these criteria. These methods can help determine the most appropriate strategy for medical waste disposal by evaluating environmental, economic, social, and technological factors simultaneously.

The application of MCDM in medical waste management has secured a key place in the literature in the last decade. Most of the studies focus on determining the most appropriate option by evaluating medical waste disposal methods in line with various criteria.

In terms of medical waste disposal methods, the four most frequently evaluated methods are Incineration, Steam Sterilization/Autoclave, Microwave Sterilization, and Landfill [3,4,5]. Additionally, there are studies examining a broader set of alternatives, such as chemical disinfection, plasma pyrolysis, radiation, and reverse polymerization [6].

In terms of criteria, most studies group evaluation criteria into four main dimensions: economic, environmental, technical, and social [7,8,9]. In these studies, capital cost, annual operating cost, and net cost per ton were considered as economic criteria [5,6,8]; waste residuals, release with health effects, and environmental impact were considered as environmental criteria [5,10]; reliability, treatment effectiveness, treatment system capacity, and infrastructure requirement were considered as technical criteria [5,6]; and public acceptance was considered as social criteria [10]. In addition, ergonomic criteria were also used in another study [7] and this study made an original contribution to the literature by including physical workload requirement, personal protective equipment requirement, the impact of occupational hazards, and psychological impact as ergonomic criteria.

Various fuzzy set extensions have been used in previous studies to model the uncertainty and hesitancy in decision-makers’ evaluations. For example, classical/triangular fuzzy sets have been used in traditional MCDM approaches and model fundamental uncertainties [11]. Interval-valued fuzzy sets (IVFS) and linguistic approaches such as the Interval 2-tuple Linguistic (ITL) approach [3] and ITI-TOPSIS [8] process linguistic information without distortion and allow decision-makers to use different granular linguistic term sets (Multi-Granularity Linguistic Distribution Assessment (MGLDA)). Intuitive fuzzy sets (IFS) simultaneously consider membership and non-membership degrees. It forms the basis of methods such as IF-EDAS and intuitionistic hesitant fuzzy sets (IHF-MULTIMOORA) [10,12]. Pythagorean fuzzy sets (PFS) extend IFS by modeling a larger domain. PF was used in the SWARA-ARAS integration [13]. Fermatean fuzzy sets (FFS) are more general than PFS. Interval-valued fermatean fuzzy sets (IVFFS) have been used with Entropy, PIPRECIA, and MARCOS methods to evaluate medical waste disposal options [14]. Spherical fuzzy sets (SFS/T-SFS) offer the ability to independently model membership, non-membership, and hesitation degrees. It has been used in advanced methodologies such as SF CRITIC-WASPAS [5] and T-SF CRITIC-MAUT [15]. Hesitant fuzzy linguistic term sets (HFLTS) offer comparative and rich linguistic term sets to capture the hesitant nature of decision-makers (MC-HFLTS and MAIRCA/MABAC) [7].

When we examine the integrated (hybrid) MCDM methodologies used in studies on disposal method selection, different methods have been employed for determining the weights. In the 2-tuple DEMATEL and fuzzy MULTIMOORA methods, 2-tuple DEMATEL was used to measure the interdependence between the criteria and the alternatives were ranked using fuzzy MULTIMOORA [3]. IHF-MULTIMOORA extended this method with intuitive hesitant fuzzy sets [12]. In the SWA-RA-ARAS hybrid method, SWARA is used to calculate the subjective weights of the criteria and ARAS is used for ranking in the context of Pythagorean fuzzy sets [13]. In the CRITIC-WASPAS/MAUT hybrid method, the CRITIC method is used to calculate the objective weights and these weights are integrated to perform ranking in a spherical fuzzy environment with WASPAS [5] or MAUT [15]. In the Entropy, PIPRECIA, MARCOS (IVFFS) hybrid method, the weights obtained using IVFFS-Entropy and IVFFS-PIPECIA are used to rank the alternatives with the IVFFS-MARCOS method [14]. When examining hybrid methods based on distance and ideal solutions, the EDAS method was adapted for the evaluation of the methods in intuitionistic fuzzy sets (IF-EDAS) [10] and MGLDA [16] environments. The study by Ju et al. (2020) used the Dice similarity criterion to determine expert weights separately for each criterion [16]. The fuzzy VIKOR method [3] was used to rank the alternative methods. The ranking was performed using the Interval 2-tuple Induced TOPSIS (ITI-TOPSIS) method [8]. Considering the fuzzy RVIKOR method [6], 11 alternatives were ranked using RVIKOR and criteria weights were determined using F-AHP. In the COVID-19 waste management implemented in Sivas, Fuzzy DBM and F-PSI methods were integrated for criteria prioritization, while the F-CRADIS method was integrated for ranking [11]. In a study using MAIRCA and MABAC methods combined with MC-HFLTS, disposal methods were selected based on hierarchical criteria [7].

When we look at the application areas of waste disposal method selection in the literature, we see that multi-criteria methods have generally been applied in regions such as China (Shanghai, Beijing), Turkey (Sivas, Erzurum), England, Libya, and Iran (Tehran, Qazvin). Considering the disposal method according to the application areas, the study in Erzurum [7], the Fuzzy TOPSIS in Shanghai [8], the ITL-MULTIMOORA [3] PF-SWARA-ARAS [13] studies, and the F-CRADIS study in Sivas [11] put steam sterilization/autoclave in the top place. In the Fuzzy AHP-VIKOR method applied in Tehran, microwave had the highest priority, while autoclave came in second [6]. In studies conducted in Qazvin using the AHP-TOPSIS method [17] and in Libya using the Grey Systems Theory method [18], microwave was also identified as the best method. In a study conducted in the UK using the AHP method, deep landfill was selected as the best disposal method.

When we examine the selection of medical waste disposal methods based on the MCDM methods used, most traditional fuzzy and intuitive fuzzy MCDM models identified steam sterilization/autoclave as the most suitable alternative. A comparative analysis using MC-HFLTS with MAIRCA and MABAC methods by Adar and Delice (2019), interval-valued fuzzy DE-MATEL and fuzzy TOPSIS methods by Li et al. (2020), and Pythagorean fuzzy COPRAS methods by Chaurasiya and Jain (2022) showed that steam sterilization was the best alternative in all methods, with the same ranking [7,19,20]. However, the incineration method emerged as the most acceptable action in methods such as global fuzzy CRITIC-WASPAS [5], intuitionistic fuzzy sets [21], and IVFF MARCOS-PIPRECIA [14]. Furthermore, the recycling method ranked first in a study integrating the AROMAN and CRITIC methods with Q-rung fuzzy [22].

Studies examining the importance levels of criteria generally show that environmental and health impacts are more important than economic criteria such as cost. For example, in the IF-EDAS model, public acceptance was identified as the most important criterion, followed by release with health effects as the second-most important criterion [10]. Beheshtinia et al. (2023) determined the order of importance of environmental, economic, technical, and social criteria [6].

A detailed summary of the literature review is provided in

Table A1. Summary of literature review.

Authors	Methods	Main Criteria	Alternative Disposal Methods	Results	Country/ Location
Liu et al. (2015) [3]	A novel Hybrid 2-tuple DEMATEL-Fuzzy MULTIMOORA	Economic, Environmental, Technical, Social	Incineration, Steam Sterilization, Microwave, Landfill	Steam Sterilization	Shanghai, China
Kalhor et al. (2016) [17]	AHP-TOPSIS	Cost, solid residuals and environmental impacts, water residuals and environmental impacts, air residuals and environmental impacts, odor, release with health effects, reliability, treatment effectiveness, level of automation, occupational hazards occurrence frequency, public acceptance obstacles, land requirement	Irradiation, Microwave, Steam Sterilization, Chemical disinfection, Sanitary Landfill, Incineration	Irradiation and Microwave.	Qazvin
Lee et al. (2016) [31]	AHP	Legal and compliance, guidelines, environmental, sustainability and carbon, cost	Deep Landfill, Incineration, Alternative Technology (autoclaving)	Deep Landfill	East Midlands region of England
Lu et al. (2016) [8]	2-tuple TOPSIS	Cost per ton, waste residues, release with health effects, reliability, treatment efficiency, public acceptance	Incineration, Steam Sterilization, Microwave, Landfill	Steam Sterilization	Shanghai, China
Shi et al. (2017) [4]	Cloud model-MABAC method	Cost per ton, waste residues, release with health effects, reliability, treatment efficiency, public acceptance	Incineration, Steam Sterilization, Microwave, Landfill	Steam Sterilization	Shanghai, China
Adar and Delice (2019) [7]	Hesitant fuzzy linguistic term set; MABAC, MAIRCA, VIKOR, TOPSIS	Economic, social, environmental, technical, ergonomic	Incineration, Steam Sterilization, Microwave, Landfill	Steam Sterilization	Erzurum, Turkey
Badi et al. (2019) [18]	Grey systems theory	Waste residues, release with health effects, treatment efficiency, net cost per ton, public acceptance	Incineration, Steam Sterilization, Microwave, Landfill	Microwave	Libya
Geetha et al. (2019) [12]	Intuitionistic hesitant fuzzy set, MULTIMOORA, TOPSIS	Automation, safety systems, cost, noise, stink, solid dregs, water residues, air pollution, land requirement, workers	Incineration, Landfill, Deep burial, Autoclave, Microwave, Chemical disinfection	Deep burial	General
Ju et al. (2020) [16]	EDAS (multi-granular linguistic distribution assessment)	Public attitude, processing cost, waste residuals, health effects of emissions, treatment effectiveness, advancement of processing technology	Incineration, Steam Sterilization, Landfill, Chemical disinfection	Steam Sterilization	Beijing, China
Li et al. (2020) [19]	Interval-valued fuzzy DEMATEL-TOPSIS	Economic sustainability, environmental sustainability, social sustainability, technical sustainability, resource sustainability	Incineration, Steam Sterilization, Landfill, Microwave	Steam Sterilization	Beijing, China
Mishra et al. (2020) [10]	Intuitionistic fuzzy EDAS	Cost, waste residuals, release with health effects, reliability, treatment effectiveness,	Microwave, Incineration, Steam Sterilization, Landfill	Steam Sterilization	HimachalPradesh, India
Makan and Fadili (2021) [9]	PROMETHEE	Environmental, financial/economic, social, technical	Municipal Landfill, Inertization, Encapsulation, Rotary kiln	Rotary kiln	General
Chaurasiya and Jain (2022) [20]	Pythagorean fuzzy COPRAS	Cost, disposal cost, energy consumption, treatment effectiveness, level of automation, need for skilled operators, public acceptance, land requirement	Steam Sterilization, Microwave, Plasma pyrolysis, Chemical disinfection, Incineration	Steam Sterilization	India
Rani et al. (2022) [13]	Pythagorean fuzzy SWARA-ARAS	Cost, waste residuals, energy consumption, treatment effectiveness, level of automation, need for skilled operators, public acceptance, land requirement	Steam Sterilization, Microwave, Plasma pyrolysis, Chemical disinfection, Incineration	Steam Sterilization	India
Salimian and Mousavi (2022) [21]	Intuitionistic Fuzzy Sets (IFS)	Cost of net per ton, residuals of the waste, health effect release, reliability, effectiveness of the treatment, admission of the society	Incineration, Steam Sterilization, Microwave, Landfill	Incineration	Shanghai, China
Anafi et al. (2023) [15]	T-Spherical Fuzzy CRITIC-MAUT	Cost, waste residuals, release with health effects, energy consumption, reliability, volume reduction, treatment effectiveness, public acceptance	Steam Sterilization, Incineration, Chemical disinfection, Microwave, Landfill disposal	Landfill disposal	China
Beheshtinia et al. (2023) [6]	Fuzzy AHP-Fuzzy VIKOR	Economic, Environmental, Technical, Social	Sanitary Landfill, Incineration, Microwave, Sterilization by autoclave, Chemical disinfection, Radiation, Treatment by NEWater process, Encapsulation, Compaction, Reverse polymerization, Plasma pyrolysis	Microwave, Sterilization by autoclave and Reverse polymerization	Tehran
Menekşe and Akdağ (2023) [5]	Spherical Fuzzy CRITIC-WASPAS	Waste residuals, infrastructure requirement, annual operating cost, treatment systems capacity, reliability, health effects, treatment efficiency, human resource requirement	Chemical disinfection, Encapsulation, Landfill, Electromagnetic Wave Sterilization, Incineration	Incineration	General
Anjum et al. (2024) [22]	Q-rung Fuzzy AROMAN-CRITIC	Cost, environmental impact, technological feasibility, compliance with regulations, safety and health	Incineration, Microwaving, Autoclaving, Landfilling, Recycling	Recycling	General
Kirişçi (2024) [14]	Interval-Valued Fermatean Fuzzy-Entropy, MARCOS, PIPRECIA	Waste residuals, infrastructure requirement, annual operating cost, treatment systems capacity, reliability, health effects, treatment efficiency, human resource requirement	Incineration, Encapsulation, Landfill, Electromagnetic Wave Sterilization, Disinfection with chemicals	Incineration	General
Demir and Moslem (2024) [11]	F-MCDM Hibrit (F-DBM, F-PSI, F-CRADIS)	Environmental, economic, technology, social	Incineration, Chemical disinfection, Autoclave, Encapsulation, Distillation, Ozonation, UV ray exposure, Chlorination, Rendering inert	Autoclave	Sivas, Turkey

.

As a result, many researchers have carried out numerous studies in which different MCDM methods have been used to solve decision-making problems related to the selection of medical waste disposal methods. Research in recent years has not only selected the best disposal method but also analyzed this process under more complex and realistic scenarios. This study applied the MCDM methods to select the most appropriate disposal method at a university hospital in Adana province, a metropolitan city in Turkey. As a major metropolitan city in Turkey, Adana produces a significant amount of medical waste with its dense population and advanced health infrastructure. Therefore, the selection of the most appropriate medical waste disposal method at hospitals in Adana is vital to establish an effective and sustainable waste management system. However, this is a very difficult decision to make when traditional methods are employed. The large number of criteria affecting the decision-making process prevents a single method from being the best for all criteria. This study was conducted to eliminate the said complexity and studied 12 criteria (initial investment cost, operating cost, environmental risk, emissions, air pollution control, feasibility, capacity, efficiency, types of waste that can be disposed of, energy recovery, waste residues, public acceptance) and four alternatives (incineration, steam sterilization, microwave, landfill) based on the literature and the expert opinions. While the CRITIC method was applied for weighting the aforementioned criteria, TOPSIS, PROMETHEE, and EDAS methods were applied for evaluating the alternatives. As a result of the methods, incineration was selected as the most appropriate alternative for the university hospital, thanks to its high efficiency.

Original Contributions of Research

The studies in the literature have focused on the application of different MCDM methods and the development of integrated models, demonstrating that the use of MCDM methods in the selection of medical waste disposal methods has become widespread, with each method offering unique results by emphasizing different criteria. However, there is no comprehensive study that addresses this issue in Turkey, particularly in Adana province. This paper aims to fill this gap and offer an objective and systematic decision-making process appropriate to the specific conditions of a university hospital in Adana. The application of CRITIC, TOPSIS, PROMETHEE, and EDAS methods, which were previously used for similar purposes, increased the reliability and up-to-dateness of the results obtained. Furthermore, the study will help healthcare institutions to make more scientific and systematic decisions in medical waste management.

The phases of this original study are as follows: Section 2 provides information about the methods used, data collection and identification, while Section 3 focuses on the weighting of the criteria, evaluation of the alternatives according to the methods used, and sensitivity analysis. Section 4 offers an interpretation of the results obtained and makes suggestions for future studies.

2. Materials and Methods

2.1. Methodology of Research

This study consists of five main sections: Firstly, the purpose and limits of the study were determined. The problem of selecting the most appropriate method for the disposal of medical wastes at a university hospital located in Adana province, a metropolitan city in Turkey, is addressed. Secondly, for the selection of medical waste disposal methods, 12 criteria were determined and defined by considering both the literature and the expert opinions, and four alternative disposal methods suitable for the medical wastes at the hospital were defined. Thirdly, the defined criteria were ranked according to their importance by using the CRITIC method, followed by the ranking of the disposal methods using TOPSIS, PROMETHEE, and EDAS methods and the testing of the stability of the methods used for selection under seven different scenarios. Fourthly, the ranking of the alternatives is summarized and, finally, the results obtained are interpreted and recommendations for further studies are presented. Figure 1 shows the steps of the research in detail.

2.2. Data Collection

2.2.1. Defining the Method Selection Problem for Medical Waste Disposal

In this study, incineration, steam sterilization, microwave, and landfill were determined as the disposal methods appropriate for the hospital in the light of the information from the experts at the university hospital in Adana province.

One of the most common methods historically used for the disposal of medical waste is incineration. Although incineration is considered an effective method because it significantly reduces the volume of waste and destroys pathogens, it also brings serious environmental problems. Toxic gases such as dioxins and furans from the incineration process may cause air pollution and acid rain, while ashes containing heavy metals are also categorized as hazardous wastes. This has made the installation and operation of incineration plants more complex and costly. Methods other than incineration include technologies such as autoclaving, microwave irradiation, and chemical disinfection. These methods usually aim at the sterilization of wastes, which can then be disposed of in landfills like municipal wastes. However, each of these methods has its own technical, economic, and environmental limitations. For instance, autoclaving may not be sufficient for large volumes of wastes, while chemical disinfection methods may result in the generation of secondary chemical wastes.

As a result, the hierarchical structure of the disposal method selection problem considered for the university hospital in Adana is provided in Figure 2.

2.2.2. Defining Criteria for Selection of Medical Waste Disposal Method

Medical waste management is addressed from a broader perspective by associating it with the Sustainable Development Goals (SDGs). Studies are no longer limited to basic criteria such as cost and environmental impact. In addition, researchers include criteria such as social (local community acceptance, employment generation) and technological criteria (maintenance costs, technological lifetime, need for residential space) in the decision-making process. It is observed that regional and hospital-based studies have been carried out, particularly in Turkey.

This study has determined the criteria for the selection of medical waste disposal methods based on the opinions of experts from the university hospital. In addition, studies in the literature were also considered. This section explains the twelve criteria (initial investment cost, operating cost, environmental risk, emissions, air pollution control, feasibility, capacity, efficiency, types of waste that can be disposed of, energy recovery, waste residues, public acceptance) affecting the selection of medical waste disposal methods.

• Initial Construction Cost (ICC)

This covers all costs incurred during the establishment phase, i.e., before the operation phase, of the medical waste processing and disposal facility. Examples include the cost of vehicles and equipment, facility set-up and real estate costs, legal permits and licenses, personnel and training costs, and other initial costs. It is desirable that this criterion be at a minimum.

• Operating Costs (OC)

These are costs incurred during the implementation of the methods applied for the processing and disposal of medical wastes. Examples include, inter alia, collection and sorting costs, personnel costs, consumables, training costs, chemicals, materials, and technical consultancy services. In addition to these costs, the costs caused by the maintenance of the applied methods in case of mechanical failure and damage can also be added. It is desirable that this criterion be at a minimum.

• Environmental Risk (Public health and safety) (ER)

This criterion expresses the risks such as air pollution and toxic emissions, water and soil pollution, ecosystem and biological threats, fires, and spread of diseases that may be caused by the methods applied for the processing and disposal of medical wastes. It is desirable that this criterion be at a minimum.

• Emissions (E)

This criterion expresses the risks such as air pollution and toxic emissions, water and soil pollution, ecosystem and biological threats, fires, and spread of diseases that may be caused by the methods applied for the processing and disposal of medical wastes. Examples include air pollution emissions, emissions during sterilization, and storage phases. It is desirable that this criterion be at a minimum.

• Air Pollution Control (APC)

This is a by-product of the disposal method applied and also shows the control of the formation of flue gases that may cause air pollution. High process temperatures and abundant oxygen lead to emissions of fly ash and toxic nitrogen components. This obliges municipalities to use efficient technologies and costly flue gas treatment systems. Maximum control ensures the fulfilment of legal and environmental requirements.

• Feasibility (F)

This is a concept that expresses how successful and sustainable a waste management system is in technical, economic, environmental, and legal terms. It shows the ability of disposal methods to fulfil their desired functions sufficiently. It is desirable that this criterion be at a maximum.

• Capacity (C)

In terms of disposal methods, this refers to the amount of waste that can be disposed of per unit time. It is desirable that this criterion be at a maximum.

• Productivity (P)

This expresses the rate of reduction in the volumetric and weight basis of the wastes in the disposal methods in question. It can also be referred to as waste minimization or waste utilization potential. It is desirable that this criterion be at a maximum.

• Types of Waste That Can Be Disposed of (TOW)

This refers to the types of waste that the disposal facility can accept and process. For instance, all wastes can be stored in sanitary landfill, medicines cannot be incinerated, and wastes must be biodegradable in biological disposal methods. It is desirable that this criterion be at a maximum.

• Energy Recovery (ERE)

This refers to the amount of potential energy that can be recovered in the disposal method. It is the capture of thermal energy released during the incineration of waste or other thermal treatment processes and its conversion into a useful form of energy. It is desirable that this criterion be at a maximum.

• Waste Residue (WR)

This refers to the criterion to be considered to ensure the harmful effects of the residues to be formed during the disposal of wastes on the environment are minimal. It is desirable that this criterion be at a minimum.

• Public Acceptance (PA)

Public participation should be ensured by taking into consideration the impacts that may be caused in the environment during the disposal of wastes under the EIA report. Public acceptance is also taken as a criterion [23]. It is desirable that this criterion be at a maximum.

2.2.3. Defining the Alternative Medical Waste Disposal Methods

In Türkiye, disposal methods of medical waste have been determined within the scope of the Waste Management Regulation published by the Ministry of Environment and Urbanization [24].

Four of the methods are applied for the disposal of medical wastes at the university hospital in Adana province. Incineration, steam sterilization, microwave irradiation, and landfill.

Incineration; Incineration is a dry oxidation process carried out at high temperatures that converts organic and flammable waste into inorganic and non-combustible materials, resulting in a significant reduction in waste volume and weight. Incineration is generally used to process waste that cannot be recycled, reused, or disposed of in landfills. Incineration involves the chemical and physical breakdown of organic material through the processes of combustion, pyrolysis, or gasification. The combustion of organic materials produces numerous gaseous emissions, including steam, carbon dioxide, nitrogen oxides, particulate matter, some toxic substances, and ash.

There are three general types of incineration technologies commonly used for the treatment of medical waste [25]:

• Dual-chamber starved-air incinerators: This method consists of a primary and a secondary chamber. The primary chamber: While decomposing waste through an oxygen-poor, medium-temperature combustion process (800–900 °C), it produces solid ashes and gases. The secondary chamber performs high-temperature incineration (1100–1600 °C) using excess air to minimize gases, smoke, carbon monoxide, and odors. These facilities are designed to incinerate infectious medical waste.
• Multiple Chamber Incinerators: These types include in-line incinerators and retort incinerators and are used for pathological waste. These facilities operate in excess-air mode and use supplemental fuel to reach temperatures of approximately 800–1000 °C.
• Rotary Kilns: These facilities have a rotating furnace and a post-combustion chamber. They can typically reach temperatures between 900 °C and 1200 °C. Because they can reach high temperatures capable of degrading genotoxic substances and heat-resistant chemicals, they can be designed specifically for the incineration of chemical waste. When appropriate combustion conditions and effective flue gas cleaning systems are used, they can be used in large-scale regional medical waste incineration facilities or for the disposal of other hazardous/toxic wastes.

Advantages and Disadvantages of Incineration: Some of the advantages of incineration include significantly reducing waste volume and weight. The thermal degradation process eliminates potentially hazardous microorganisms, reducing the risk of disease transmission to humans and the environment. Incineration can dispose of a wide variety of waste types (e.g., sharps, pharmaceutical residues, infectious contaminants, pathological waste, and other biological waste categories). Furthermore, the ability to recycle the energy released from incineration and use it as a renewable energy source is a significant advantage. On the other hand, if improper methods are not used, advanced and expensive flue gas treatment systems are required because they can lead to emissions such as dioxins, furans, heavy metals, particulate matter, and greenhouse gases that are harmful to human health and the environment. Furthermore, high investment and operating costs are among the main disadvantages of incineration [22].

Steam Sterilization; Steam sterilization is a system consisting of a metal pressure vessel that effectively sterilizes infectious waste using high-pressure saturated steam [22,25]. It can inactivate a variety of biological wastes, including cultures, materials contaminated with blood and limited amounts of liquid, laboratory waste (excluding chemical waste), soft waste (gauze, bandages, gowns, bedding, etc.), sharps, insulation, and surgical waste. However, this method is not suitable for large anatomical parts (body parts). Removal of air from the interior is critical for the device’s efficient operation. Waste treatment autoclaves must purify the air removed at the beginning of the process to prevent the release of pathogenic aerosols; this is typically achieved by steam purification or passing the air through a high-efficiency particulate air (HEPA) filter before release [25]. Steam sterilization destroys pathogens by utilizing the heating and penetration properties of steam and requires a continuous supply of electricity and water to generate the necessary steam [22].

Advantages and Disadvantages of Steam Sterilization: Steam sterilization offers significant advantages, including reducing the risk of infection by effectively destroying bacteria, viruses, and spores; simplifying storage and transportation by reducing waste volume; and lowering environmental emissions compared to methods such as incineration. However, significant disadvantages include the need for an uninterrupted water and electricity supply to maintain the process and the fact that not every type of waste, especially chemically contaminated or some hazardous waste, is suitable for steam sterilization [22].

Landfill; Landfill is a method used to dispose of waste by burial in a controlled manner, particularly when healthcare facilities must dispose of their waste directly before treatment [22,25]. In less-developed areas, where treatment is not available, directing waste to sanitary landfills is a common practice and carries a much lower risk of infection than methods such as open burning or random dumping. Controlled landfills utilize improved working practices and designs to reduce environmental and health impacts and include progressive engineering levels such as controlled dumping, engineered landfill, and sanitary landfill. These sites are designed to reduce waste leaching into soil, surface water, and groundwater, to prevent pests, and to limit odor and emissions [25]. Landfill can be used for the disposal of non-infectious and non-hazardous waste as well as some types of sharps and needles [22].

Landfill Advantages and Disadvantages: Landfilling offers the advantages of being an accessible and easy-to-implement disposal method thanks to its widespread use. It reduces the risk of environmental contamination by safely containing waste. It also offers long-term storage at a lower cost compared to other methods. However, improper storage of biomedical waste poses a significant disadvantage: the leakage of harmful substances into the environment, threatening human health and ecosystems. Therefore, strict legal frameworks and effective waste separation practices are necessary [22].

Microwave; Microwave technology is a disinfection process that uses moist heat and steam generated by rapidly heating water in waste with microwave energy at a frequency of approximately 2450 MHz [22,25]. Systems typically include a microwave generator consisting of magnetrons, a processing chamber, an automatic loading system, a hopper, grinders, conveyor screws, a steam generator, a discharge screw, a secondary shredder, a HEPA filter, and a microprocessor-controlled structure [25]. Waste types that can be processed with microwaves include sharps, cultures, materials contaminated with blood and body fluids, surgical and isolation waste, laboratory waste, and soft waste (gauze, bandages, gowns, bedding. This method provides rapid and effective disinfection due to the inherent heat generation of the waste; however, volatile organics, chemotherapeutic waste, hazardous chemicals, and radioactive waste should not be processed with microwaves [22,25]).

Advantages and Disadvantages of Microwaving: Microwaving offers advantages such as rapid and effective disinfection, based on the generation of heat within the waste, and simplifies storage and transportation processes by reducing waste volume. However, for the system to operate efficiently, a reliable power supply and appropriate infrastructure are required; careful waste preparation and processing are also essential to ensure consistent heating and disinfection. Furthermore, a significant limitation of the method is that some types of chemical, pharmaceutical, or radioactive waste cannot be processed with microwaves [22].

2.2.4. Proposed Model for Medical Waste Disposal Method Selection Problem

The process of the proposed MCDM model for the selection of a medical waste disposal method at a university hospital in Adana province is provided in Figure 3. In the proposed model, the criteria were initially selected by considering the opinions of experts and the literature, followed by the forming of the decision matrix. Then, the criteria weights were determined using the CRITIC method, followed by the ranking of alternative medical waste methods using TOPSIS, PROMETHEE, and EDAS methods. In the final phase of the study, sensitivity analysis was performed to test the reliability of the results.

2.3. CRITIC Method

The CRITIC method was first introduced by Diakoulaki in 1995 [26]. It is generally used in the evaluation of the criteria in the case of MCDM problems. This method takes into account the standard deviations of the criteria as well as the correlation among them. In general, the CRITIC method consists of the steps shown in Figure 4 [27]. A detailed explanation of each step is given in

Table A2. Stages of the CRITIC method.

CRITIC Method
Steps	Formula	Explanation
Step 1: Identifying the problem:	-	The structure of the addressed MCDM problem is revealed. Both its scope and purpose are determined.
Step 2: Determining criteria and alternatives:	-	Criteria suitable for the set purpose as well as alternatives for evaluation are determined.
Step 3: Creating the decision matrix:	X = $x_{ij}$ =$\begin{array}{c}{A}_1\\ A_2\\ .\\ .\\ A_m\end{array}$ $\left[\begin{array}{c}{x}_{11} x_{12}\dots x_{1n}\\ x_{21} x_{22}\dots x_{2n}\\ .\\ .\\ x_{m1} x_{m2}\dots x_{mn}\end{array}\right]$ (A1)	$\mathrm{The} \mathrm{decision} \mathrm{matrix} \mathrm{consisting} \mathrm{of} \mathrm{m} \mathrm{alternatives} \mathrm{and} \mathrm{n} \mathrm{criteria} \mathrm{in} \mathrm{Equation} \left(\mathrm{A}1\right) \mathrm{is} \mathrm{created}. x_{ij}$ refers to the value of alternative i. according to criterion j.
Step 4: Creating the normalized decision matrix:	$r_{ij}$$= {\displaystyle \frac{x_{ij}-x_j^{min}}{x_j^{max}-x_j^{min}}}$ j = 1, 2,…, n (A2) $r_{ij}$ = ${\displaystyle \frac{x_j^{max}-x_{ij}}{x_j^{max}-x_j^{min}}}$ j = 1, 2,…, n (A3)	After the decision matrix normalized in the range [0,1]. Equation (A2) is used in the maximization criteria, and Equation (A3) is used for the criteria in the minimization criteria.
Step 5: Creating the correlation matrix:	${\rho }_{jk}$$= {\displaystyle \frac{{\sum }_{i=1}^m\left(r_{ij} - \overline{r_j}\right) . (r_{ik}- \overline{r_k})}{\sqrt{{{\sum }_{i=1}^m (r_{ij- }\overline{r_j})}^2. { {\sum }_{i=1}^m(r_{ik }- \overline{r_k})}^2}}}$ (A4)	The relationship coefficient matrix between the evaluation criteria is created by the formula provided in Equation (A4).
Step 6: Obtaining  ${\mathit{C}}_{\mathit{j}}$values:	$C_j= {\sigma }_j . {\sum }_{k=1}^n(1-{\rho }_{jk})$ (A5) ${\sigma }_j= \sqrt{{\displaystyle \frac{{{\sum }_{i=1}^m (r_{ij- }\overline{r_j})}^2}{m-1}}}$ (A6)	$\mathrm{At} \mathrm{this} \mathrm{step}, C_j$ that the total information value of each criterion j is calculated by the formula in Equations (A5) and (A6).
Step 7: Calculating criterion weights:	$w_j= {\displaystyle \frac{C_j}{{\sum }_{k=1}^n{C}_j}}$ (A7)	$\mathrm{Finally}, \mathrm{the} \mathrm{weight} \mathrm{values} w_j$ of each criterion, are calculated as in Equation (A7).

.

2.4. TOPSIS Method

The TOPSIS method is an MCDM method first developed by Hwang and Yoon in 1981 [28]. This method takes into account the proximity of alternatives to the positive ideal solution and their distance from the negative ideal solution. By comparing the distances to these ideal solutions, the ranking of the alternatives is determined. The methodology of the TOPSIS method consists of the steps in Figure 5 [27]:

A detailed explanation of each step is given in

Table A3. Stages of the TOPSIS method.

TOPSIS Method
Steps	Formula	Explanation
Step 1: Creating the decision matrix:	$\mathrm{A} = a_{ij}$$= \left[\begin{array}{c}{a}_{11} a_{12}\dots a_{1n}\\ a_{21} a_{22}\dots a_{2n}\\ .\\ .\\ a_{m1} a_{m2}\dots a_{mn}\end{array}\right]$ (A8)	Firstly, decision matrix A is created by decision-makers in Equation (A8).
Step 2: Creating the standard decision matrix:	$x_{ij}= {\displaystyle \frac{a_{ij}}{\sqrt{{\sum }_{i=1}^m{a_{ij}}^2}}} \left(i=\mathrm{1,2},\dots ,m j=\mathrm{1,2},\dots n\right)$ (A9) X = $x_{ij}$ = $\left[\begin{array}{c}{x}_{11} x_{12}\dots x_{1n}\\ x_{21} x_{22}\dots x_{2n}\\ .\\ .\\ x_{m1} x_{m2}\dots x_{mn}\end{array}\right]$ (A10)	$\mathrm{After} \mathrm{each} a_{ij}$ value is standardized in the range [0,1] by the formula in Equation (A9) and standard decision matrix is obtained in Equation (A10).
Step 3: Creating the weighted standard decision matrix:	$W_{ij}$ = $\left[\begin{array}{ccc}{w}_1{x}_{11}& w_2{x}_{12}& \dots \\ \begin{array}{c}{w_{1}x}_{21}\\ ⋮\end{array}& \begin{array}{c}{w_{2}x}_{22}\\ ⋮\end{array}& \begin{array}{c}\dots \\ \dots \end{array} \\ w_1{x}_{m1}& {w_{2}x}_{m2}& \dots \end{array}\begin{array}{c}{w_{n}x}_{1n}\\ \begin{array}{c}{w_{n}x}_{2n}\\ ⋮\end{array}\\ w_n{x}_{mn}\end{array}\right]$ (A11)	$\mathrm{Weighted} \mathrm{decision} \mathrm{matrix} (W_{ij}$) shown in Equation (A11) is obtained by multiplying the values in the standard decision matrix with the weights.
Step 4: Determining positive and negative ideal solution:	$A^{+}= \left\{ \left(\left(i\right)max{ w}_{ij}|j ϵ J\right), \left(\left(i\right)\mathit{min}{w}_{ij}|j ϵ J*\right)\right\}$ $A^{+}= \left\{w_1^{+},w_2^{+},\dots .,w_n^{+} \right\}$ (A12) $A^{-}= \left\{ \left(\left(i\right)min{ w}_{ij}|j ϵ J\right), \left(\left(i\right)\mathit{max}{w}_{ij}|j ϵ J*\right)\right\}$ $A^{-}= \left\{w_1^{-},w_2^{-},\dots .,w_n^{-} \right\}$ (A13)	Positive and negative ideal solution values are determined from the matrix Wij created in the previous step. The positive ideal solution value is calculated as in Equation (A12). The negative ideal solution value is calculated as in Equation (A13).
Step 5: Determining distances to the positive and negative ideal solution:	$U_i^{+}=\sqrt{{\sum }_{j=1}^n{\left({ w}_{ij}- w_j^{+}\right)}^2}$ (A14) $U_i^{-}=\sqrt{{\sum }_{j=1}^n{\left({ w}_{ij}- w_j^{-}\right)}^2}$ (A15)	The distance to the positive ideal solution is calculated as in Equation (A14) while the distance to the negative ideal solution is calculated as in Equation (A15) using the Euclidean distance approach.
Step 6: Ranking alternatives by determining distances to the ideal solution:	$C_i^{*}= {\displaystyle \frac{U_i^{-}}{U_i^{-}+ U_i^{+}}}$ (A16)	$\mathrm{The} \mathrm{distances} \mathrm{of} \mathrm{each} \mathrm{decision} \mathrm{value} \mathrm{to} \mathrm{the} \mathrm{ideal} \mathrm{solution} (C_i^{*}$$) \mathrm{are} \mathrm{calculated} \mathrm{by} \mathrm{the} \mathrm{formula} \mathrm{in} \mathrm{Equation} (\mathrm{A}16). \mathrm{If} \mathrm{the} C_i^{*}$ value is close to ‘0’, it is considered as the worst alternative, and if it is close to ‘1’, it is rated as the best alternative.

.

2.5. Promethee Method

This method was developed by Jean-Pierre Brans in 1982 and is an effective method used for ranking alternatives in MCDM problems. The method is based on pairwise comparisons of alternatives according to specified criteria, resulting in a partial ranking of alternatives by PROMETHEE I and a full (final) ranking by PROMETHEE II. The PROMETHEE method is indicated in Figure 6 [29]:

A detailed explanation of each step is given in

Table A4. Stages of the PROMETHEE method.

PROMETHEE Method
Steps	Formula						Explanation
Step 1: Creating the decision matrix and criteria weights:			Criteria				Firstly, the decision matrix is formed together with the criteria weights.
			f1	f2	…	fk
	Alternatives	A1	f1(A1)	f1(A1)	…	fk(A1)
		A2	f1(A1)	f1(A1)	…	f1(A1)
		…	…	…	…	…
		An	f1(An)	f2(An)	…	fk(An)
	Weights	wi	w1	w2		wk
Step 2: Determining preference functions of criteria:	In this step, preference functions are determined in order to show the structure of the criteria and their relationship with each other. Then, pairwise comparisons of the alternatives according to the criteria are analyzed, and the preference degree of the best alternative is determined. The chosen preference function is denoted by P and takes values between 0 and 1. There are six preference functions as regular, U-type, V-type, gradual, linear, and gaussian. These preference functions are determined by the decision-maker.
Step 3: Determining common preference functions of alternatives:	P ($A,B$) = $\left\{\begin{array}{cc}0\hfill & f\left(A\right)\le f\left(B\right)\\ P\left[f\left(A\right)-f\left(B\right)\right]\hfill & f\left(A\right) >f\left(B\right)\end{array}\right.$ (A17)						After, the common preference functions of the alternatives are calculated for each criterion. When comparing two alternatives such as a and b, common preference functions are calculated by the formula in Equation (A17).
Step 4: Determining preference indices of alternatives:	$\pi (A,B)$ = ${\sum }_{i=1}^k{(w}_i{P}_i(A,B))$ (A18)						Then, preference indices for the compared alternatives are determined using Equation (A18).
Step 5: Calculating positive and negative superiority values of alternatives:	${\varnothing }^{+}$ = ${\displaystyle \frac{1}{n-1}}\sum \pi (A,x)$ (A19) ${\varnothing }^{-}$ = ${\displaystyle \frac{1}{n-1}}\sum \pi (x,A)$ (A20)						Then, the positive and negative superiority values for the alternatives are calculated with Equations (A19) and (A20). (n is the total number of alternatives and x is the set of alternatives other than A)
Step 6: Calculating partial ranking of alternatives with PROMETHEE I:	$\mathrm{If} {\varnothing }^{+}$(A1) $>$ ${\varnothing }^{+}$(A2) and ${\varnothing }^{-}$(A1) $<$ ${\varnothing }^{-}$(A2), or $\mathrm{If} {\varnothing }^{+}$(A1) $>$ ${\varnothing }^{+}$(A2) and ${\varnothing }^{-}$(A1) $=$ ${\varnothing }^{-}$(A2), or $\mathrm{If} {\varnothing }^{+}$(A1) $=$ ${\varnothing }^{+}$(A2) and ${\varnothing }^{-}$(A1) $<$ ${\varnothing }^{-}$(A2), then alternative A1 is superior to alternative A2. $\mathrm{If} {\varnothing }^{+}$(A1) $=$ ${\varnothing }^{+}$(A2) and ${\varnothing }^{-}$(A1) $=$ ${\varnothing }^{-}$(A2), alternative A1 is the same as alternative A2. $\mathrm{If} {\varnothing }^{+}$(A1) $>$ ${\varnothing }^{+}$(A2) and ${\varnothing }^{-}$(A1) $>$ ${\varnothing }^{-}$(A2), or $\mathrm{If} {\varnothing }^{+}$(A1) $<$ ${\varnothing }^{+}$(A2) and ${\varnothing }^{-}$(A1) $<$ ${\varnothing }^{-}$(A2), then alternative A2 is superior to alternative A1.						In this step, PROMETHEE I partial ranking is determined. There are six different cases in comparisons.
Step 7: Calculating net rankings of alternatives with PROMETHEE II:	${\varnothing }^{net}$(A) $=$ ${\varnothing }^{+}$(A) $-$ ${\varnothing }^{-}$(A) (A21) $\mathrm{Where} 0 \le {\mathrm{\varnothing }}^{\mathrm{n}\mathrm{e}\mathrm{t}}$$\left(\mathrm{A}\right) \le$$1 \mathrm{and} \sum _{\mathrm{x}\in \mathrm{A}}{\mathrm{\varnothing }}^{\mathrm{n}\mathrm{e}\mathrm{t}}\left(\mathrm{A}\right)$ = 0 $\mathrm{If} {\mathrm{\varnothing }}^{\mathrm{n}\mathrm{e}\mathrm{t}}$$\left(A1\right) >$${\mathrm{\varnothing }}^{\mathrm{n}\mathrm{e}\mathrm{t}}$(A2), alternative A1 is superior to alternative A2. $\mathrm{If} {\mathrm{\varnothing }}^{\mathrm{n}\mathrm{e}\mathrm{t}}$$\left(A1\right) =$${\mathrm{\varnothing }}^{\mathrm{n}\mathrm{e}\mathrm{t}}$(A2), alternative A1 and alternative A2 are the same						In the last step, net superiority values of the alternatives are determined by PROMETHEE II using the formula in Equation (A21). They are calculated by computing the difference between the positive superiority value and the negative superiority value for each alternative.

.

2.6. EDAS Method

The EDAS method is an effective approach developed by Mehdi Keshavarz Ghorabaee et al. in 2015 [30]. In this method, positive distance (PDA) to the mean solution and negative distance (NDA) to the mean solution of alternatives takes into account. The application steps of the EDAS method are provided in Figure 7.

A detailed explanation of each step is given in

Table A5. Stages of the EDAS method.

EDAS Method
Steps	Formula	Explanation
Step 1: Creating the initial decision matrix:	X = ${\left[X_{ij}\right]}_{n\times m}$ = $\left[\begin{array}{c}{x}_{11} x_{12}\dots x_{1m}\\ x_{21} x_{22}\dots x_{2m}\\ .\\ .\\ x_{n1} x_{n2}\dots x_{nm}\end{array}\right]$ (A22)	The decision matrix consisting of m alternatives and n criteria is created as in Equation (A22).
Step 2: Creating the average solution matrix:	AV = ${\left[{AV}_j\right]}_{1\times m}$ (A23) ${AV}_j$ = ${\displaystyle \frac{{\sum }_{i=1}^n{X}_{ij}}{n}}$ (A24)	In this step, the average solution matrix (AV) containing ${AV}_j$ values for each criterion is calculated by using Equations (A23) and (A24).
Step 3: Calculating positive and negative distances from the average solution:	${PDA}_{ij}$$= {\displaystyle \frac{\mathit{max} \left(0, \left(X_{ij} - {AV}_j\right)\right)}{{AV}_j}}$ (A25) ${NDA}_{ij}$$= {\displaystyle \frac{\mathit{max} \left(0, \left({AV}_j -{ X}_{ij} \right)\right)}{{AV}_j}}$ (A26) ${PDA}_{ij}$ = ${\displaystyle \frac{\mathit{max} \left(0, \left({AV}_j -{ X}_{ij} \right)\right)}{{AV}_j}}$ (A27) ${NDA}_{ij}$ = ${\displaystyle \frac{\mathit{max} \left(0, \left(X_{ij} - {AV}_j \right)\right)}{{AV}_j}}$ (A28)	In this step, positive (PDA) and negative distance (NDA) matrices are obtained. If there is a benefit-side criterion, the values of ${PDA}_{ij}$ are calculated as in Equation (A25) and the values of ${NDA}_{ij}$ are calculated as in Equation (A26), but if there is a cost-side criterion, the values of ${PDA}_{ij}$ are calculated as in Equation (A27) and the values of ${NDA}_{ij}$ are calculated as in Equation (A28).
Step 4: Calculating weighted total positive and weighted total negative distances:	${SP}_i$ = ${\sum }_{j=1}^m{w}_j\times {PDA}_{ij}$ (A29) ${SN}_i$ = ${\sum }_{j=1}^m{w}_j\times {NDA}_{ij}$ (A30)	In this step, the total positive and total negative distance values are multiplied by the criteria weights ($w_j$) to calculate the weighted total positive values ${SP}_i$ as in Equation (A30) and weighted total negative values ${SN}_i$) as in Equation (A31).
Step 5: Normalizing weighted sum positive and weighted sum negative distance values:	${NSP}_i$ = ${\displaystyle \frac{{SP}_i}{\max \left({SP}_i\right)}}$ (A31) ${NSN}_i$ = 1 − ${\displaystyle \frac{{SN}_i}{\max \left({SN}_i\right)}}$ (A32)	In this step, the value calculated in the previous step is normalized using Equations (A31) and (A32).
Step 6: Calculating and ranking of the score of alternatives:	${AS}_i$ = ${\displaystyle \frac{1}{2}}\left({NSP}_i+{NSN}_i\right)$ (A33)	Finally, the evaluation scores (${AS}_i)$ for each alternative are calculated as in Equation (A33).

.

3. Findings and Discussion

The main objective is to make the best choice among the disposal techniques in order to minimize the damages caused by medical wastes. To realize this objective, medical waste disposal techniques applied in Turkey and in the world were examined. This study was conducted at a university hospital in Adana. The criteria and alternatives considered in the application and the relationships between them were formed by utilizing expert opinions as well as studies in the literature. In order to create decision matrices, subject matter experts were consulted for information. Information regarding the experts’ titles in the field, their experience, and their contributions to the study are shown in

Table 1. Information about experts.

Experts	Title	Experience	Contribution to the Study
Expert 1	Hospital Medical Waste Specialist	15 years	Determining criteria, alternatives and criteria weights, creating the decision matrix
Expert 2	Hospital Chief Physician	20 years	Creating the decision matrix
Expert 3	Environmental Engineer (Academic)	22 years	Creating the decision matrix

. Expert decision matrix evaluations are shown in

Table A6. Assessments of Experts.

Assessments of Expert 1
Alternatives/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
Incineration	10	7	5	10	8	1	6	10	8	9	4	10
Steam Sterilization	7	9	4	6	4	5	5	8	6	1	8	10
Microwave	8	8	4	5	5	2	4	7	4	1	7	10
Storage	4	4	6	4	2	2	7	5	1	1	10	10
Assessments of Expert 2
Alternatives/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
Incineration	8	6	3	9	10	1	8	8	10	8	3	10
Steam Sterilization	6	7	3	5	6	6	3	6	4	1	7	10
Microwave	7	7	3	4	4	2	3	6	3	1	6	9
Storage	3	3	5	3	4	3	6	4	1	1	9	10
Assessments of Expert 3
Alternatives/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
Incineration	9	5	4	8	9	1	7	9	9	10	2	10
Steam Sterilization	5	8	2	4	5	7	4	7	5	1	6	10
Microwave	6	6	2	3	3	5	2	5	2	1	5	8
Storage	2	2	4	2	3	1	5	3	1	1	8	10

. Arithmetic averaging was used to create the final decision matrix used in the methods.

Twelve criteria and four alternatives affecting the selection of the disposal method of medical wastes were discussed. The criteria include initial investment cost, operating cost, environmental risk, emissions, air pollution control, feasibility, capacity, efficiency, types of waste that can be disposed of, energy recovery, waste residues, and public acceptance. Alternatives are incineration, steam sterilization, microwave, and landfill.

3.1. Using the CRITIC Method for the Weighting of Criteria for Selection of Medical Waste Method

The study initially used the CRITIC method to weight the criteria for the selection of medical waste disposal method. The steps of the CRITIC method were calculated individually using Microsoft Excel 2016.

As a first step, the decision matrix containing the minimum and maximum directional criteria as well as alternatives considered in relation to the problem was scored on a scale of 1 to 10, resulting in the outcome in

Table 2. Decision matrix for the selection of medical waste disposal methods.

Decision Matrix
Alternatives/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
	Min	Min	Min	Min	Max	Max	Max	Max	Max	Max	Min	Max
Incineration	9	6	4	9	9	1	7	9	9	9	3	10
Steam Sterilization	6	8	3	5	5	6	4	7	5	1	7	10
Microwave	7	7	3	4	4	3	3	6	3	1	6	9
Storage	3	3	5	3	3	2	6	4	1	1	9	10

.

Then, the decision matrix was normalized using the formulas in Equations (A2) and (A3) by considering the maximum and minimum directional criteria. The resulting matrix is provided in

Table 3. Normalized decision matrix.

Normalized Decision Matrix
Alternatives/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
Incineration	0.000	0.400	0.500	0.000	1.000	0.000	1.000	1.000	1.000	1.000	1.000	1.000
Steam Sterilization	0.500	0.000	1.000	0.667	0.333	1.000	0.250	0.600	0.500	0.000	0.333	1.000
Microwave	0.333	0.200	1.000	0.833	0.167	0.400	0.000	0.400	0.250	0.000	0.500	0.000
Storage	1.000	1.000	0.000	1.000	0.000	0.200	0.750	0.000	0.000	0.000	0.000	1.000

.

As a next step, the correlation matrix between the criteria was calculated using the formula in Equation (A4) and the calculated values are provided in

Table 4. Correlation matrix between criteria.

Correlation Matrix (ρjk)
Criteria/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
ICC	1	0.617	−0.522	0.849	−0.849	0.185	−0.073	−0.929	−0.878	−0.733	−0.973	0.200
OC	0.617	1	−0.967	0.293	−0.293	−0.643	0.592	−0.593	−0.452	0.000	−0.432	0.309
ER	−0.522	−0.967	1	−0.099	0.099	0.645	−0.763	0.418	0.255	−0.174	0.313	−0.522
E	0.849	0.293	−0.099	1	−1.000	0.352	−0.555	−0.944	−0.983	−0.951	−0.925	−0.317
APC	−0.849	−0.293	0.099	−1.000	1	−0.352	0.555	0.944	0.983	0.951	0.925	0.317
F	0.185	−0.643	0.645	0.352	−0.352	1	−0.676	−0.074	−0.181	−0.617	−0.370	0.000
C	−0.073	0.592	−0.763	−0.555	0.555	−0.676	1	0.263	0.428	0.730	0.292	0.730
P	−0.929	−0.593	0.418	−0.944	0.944	−0.074	0.263	1	0.985	0.801	0.929	0.160
TOW	−0.878	−0.452	0.255	−0.983	0.983	−0.181	0.428	0.985	1	0.878	0.917	0.293
ERE	−0.733	0.000	−0.174	−0.951	0.951	−0.617	0.730	0.801	0.878	1	0.867	0.333
WR	−0.973	−0.432	0.313	−0.925	0.925	−0.370	0.292	0.929	0.917	0.867	1	−0.067
PA	0.200	0.309	−0.522	−0.317	0.317	0.000	0.730	0.160	0.293	0.333	−0.067	1

.

Then, the relationship coefficient matrix for each criterion is calculated and provided in

Table 5. Coefficient matrix of relationship between criteria.

Relationship Coefficient Matrix (1 − ρjk)
Criteria/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
ICC	0	0.383	1.522	0.151	1.849	0.815	1.073	1.929	1.878	1.733	1.973	0.800
OC	0.383	0	1.967	0.707	1.293	1.643	0.408	1.593	1.452	1.000	1.432	0.691
ER	1.522	1.967	0	1.099	0.901	0.355	1.763	0.582	0.745	1.174	0.687	1.522
E	0.151	0.707	1.099	0	2.000	0.648	1.555	1.944	1.983	1.951	1.925	1.317
APC	1.849	1.293	0.901	2.000	0	1.352	0.445	0.056	0.017	0.049	0.075	0.683
F	0.815	1.643	0.355	0.648	1.352	0	1.676	1.074	1.181	1.617	1.370	1.000
C	1.073	0.408	1.763	1.555	0.445	1.676	0	0.737	0.572	0.270	0.708	0.270
P	1.929	1.593	0.582	1.944	0.056	1.074	0.737	0	0.016	0.199	0.071	0.840
TOW	1.878	1.452	0.745	1.983	0.017	1.181	0.572	0.016	0	0.122	0.083	0.707
ERE	1.733	1.000	1.174	1.951	0.049	1.617	0.270	0.199	0.122	0	0.133	0.667
WR	1.973	1.432	0.687	1.925	0.075	1.370	0.708	0.071	0.083	0.133	0	1.067
PA	0.800	0.691	1.522	1.317	0.683	1.000	0.270	0.840	0.707	0.667	1.067	0
Total	14	13	12	15	9	13	9	9	9	9	10	10

.

As a next step, the standard deviation of each criterion was calculated using Equation (A6) and the $C_j$ values of the criteria were calculated by considering the formula in Equation (A5) as shown in

Table 6. Calculating $C_j$ values.

Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
Standard Deviation	0.361	0.374	0.415	0.380	0.380	0.374	0.395	0.361	0.370	0.433	0.361	0.433
${\mathit{C}}_{\mathit{j}}$	5.090	4.703	5.107	5.800	3.310	4.764	3.746	3.260	3.237	3.861	3.437	4.141

.

Finally, $w_j$ values, i.e., the weight values of each criterion, were calculated using the formula in Equation (A7) as shown in

Table 7. Calculating $w_j$ values and intercriteria ranking.

Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
${\mathit{w}}_{\mathit{j}}$	0.101	0.093	0.101	0.115	0.066	0.094	0.074	0.065	0.064	0.077	0.068	0.082
Ranking	3	5	2	1	10	4	8	11	12	7	9	6

.

As indicated in Table 7, Emissions (E) is identified as the most important criterion for the selection of a medical waste disposal method. Productivity (P) and Types of Waste That Can Be Disposed of (TOW) are ranked as the least important criteria.

3.2. Evaluating Alternatives for Selection of Medical Waste Method

3.2.1. Evaluating Alternative Medical Waste Disposal Methods by TOPSIS Method

Firstly, after squaring the values in the decision matrix created in Table 2, the values for each criterion were summed and their square root was taken. The calculated values are provided in

Table 8. Normalization Process.

Normalization Process
Alternatives/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
Incineration	81	36	16	81	81	1	49	81	81	81	9	100
Steam Sterilization	36	64	9	25	25	36	16	49	25	1	49	100
Microwave	49	49	9	16	16	9	9	36	9	1	36	81
Storage	9	9	25	9	9	4	36	16	1	1	81	100
$\sqrt{{\displaystyle \sum _{\mathit{i}=1}^{\mathit{m}}}{\mathit{a}}_{\mathit{i}\mathit{j}}^2}$	13.229	12.570	7.681	11.446	11.446	7.071	10.488	13.491	10.770	9.165	13.229	19.519

. Then, the value of each criterion was divided by the value in Table 8, and a standard decision matrix was established. The resulting matrix is provided in

Table 9. Standard Decision Matrix.

Standard Decision Matrix
Alternatives/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
Incineration	0.680	0.477	0.521	0.786	0.786	0.141	0.667	0.667	0.836	0.982	0.227	0.512
Steam Sterilization	0.454	0.636	0.391	0.437	0.437	0.849	0.381	0.519	0.464	0.109	0.529	0.512
Microwave	0.529	0.557	0.391	0.350	0.350	0.424	0.286	0.445	0.279	0.109	0.454	0.461
Storage	0.227	0.239	0.651	0.262	0.262	0.283	0.572	0.297	0.093	0.109	0.680	0.512
Criteria Weights	0.101	0.093	0.101	0.115	0.066	0.094	0.074	0.065	0.064	0.077	0.068	0.082

. The values added to the bottom row of this table are the criteria weight values in Table 7 as obtained by the CRITIC method. The steps of the TOPSIS method were calculated individually by using Microsoft Excel 2016.

As a next step, the criteria weight values obtained by the CRITIC method are multiplied by the values in the standard decision matrix and the weighted standard decision matrix is formed. This matrix is available in

Table 10. Weighted standard decision matrix.

Weighted Standard Decision Matrix
Alternatives/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
Incineration	0.069	0.045	0.053	0.090	0.052	0.013	0.050	0.043	0.054	0.075	0.015	0.042
Steam Sterilization	0.046	0.059	0.040	0.050	0.029	0.080	0.028	0.034	0.030	0.008	0.036	0.042
Microwave	0.053	0.052	0.040	0.040	0.023	0.040	0.021	0.029	0.018	0.008	0.031	0.038
Storage	0.023	0.022	0.066	0.030	0.017	0.027	0.043	0.019	0.006	0.008	0.046	0.042

.

Then, positive and negative ideal solutions are calculated. For the positive ideal solution, the largest value in the column is taken if the objective of the criterion is maximum and the smallest value in the column is taken if it is minimum. For the negative ideal solution, the smallest value in the column is taken if the objective of the criterion is maximum and the largest value in the column is taken if it is minimum. As seen in

Table 11. Determining positive and negative ideal solutions.

Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
Positive Ideal Solution	0.023	0.022	0.040	0.030	0.052	0.080	0.050	0.043	0.054	0.075	0.015	0.042
Negative Ideal Solution	0.069	0.059	0.066	0.090	0.017	0.013	0.021	0.019	0.006	0.008	0.046	0.038

, the positive ideal solution desires that the initial investment cost, operating cost, environmental risk, emissions and waste residues are minimum, while air pollution control, feasibility, capacity, efficiency, the types of waste that can be disposed of, energy recovery, and public acceptance are maximum. In the negative ideal solution, the minimum criteria become maximum, and the maximum criteria become minimum. Finally, positive and negative ideal solutions are calculated by considering Equations (A11) and (A12). The results of these operations are provided in Table 11.

After the positive and negative distances were determined, the distances of each value to the positive ideal solution were calculated using the formulae in Equation (A13) as shown in

Table 12. Determining distances to the positive ideal solution.

Calculation of Positive Ideal Distance
Alternatives/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA	TOTAL	${\mathit{S}}_{\mathit{i}}^{\mathit{*}}$
Incineration	0.002	0.001	0.000	0.004	0.000	0.005	0.000	0.000	0.000	0.000	0.000	0.000	0.011	0.104
Steam Sterilization	0.001	0.001	0.000	0.000	0.001	0.000	0.001	0.000	0.001	0.005	0.000	0.000	0.009	0.094
Microwave	0.001	0.001	0.000	0.000	0.001	0.002	0.001	0.000	0.001	0.005	0.000	0.000	0.011	0.107
Storage	0.000	0.000	0.001	0.000	0.001	0.003	0.000	0.001	0.002	0.005	0.001	0.000	0.013	0.114

.

Subsequently, the distances of each value to the negative ideal solution were calculated using the formulae in Equation (A14) as shown in

Table 13. Determining distances to the negative ideal solution.

Calculation of Negative Ideal Distance
Alternatives/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA	Total	${\mathit{S}}_{\mathit{i}}^{-}$
Incineration	0.000	0.000	0.000	0.000	0.001	0.000	0.001	0.001	0.002	0.005	0.001	0.000	0.011	0.103
Steam Sterilization	0.001	0.000	0.001	0.002	0.000	0.005	0.000	0.000	0.001	0.000	0.000	0.000	0.008	0.092
Microwave	0.000	0.000	0.001	0.003	0.000	0.001	0.000	0.000	0.000	0.000	0.000	0.000	0.005	0.069
Storage	0.002	0.001	0.000	0.004	0.000	0.000	0.001	0.000	0.000	0.000	0.000	0.000	0.008	0.088

.

Finally, the distances to the ideal solution were calculated and medical waste disposal methods were ranked. This evaluation is provided in

Table 14. Distances to the ideal solution and ranking of alternative methods according to TOPSIS method.

Results
Alternatives	${\mathit{C}}_{\mathit{i}}^{\mathit{*}}$	Rankings
Incineration	0.498	1
Steam Sterilization	0.493	2
Microwave	0.392	4
Storage	0.435	3

.

In conclusion, as a result of ranking by TOPSIS method, the incineration method ranks first in the selection of medical waste disposal method for the hospital, while steam sterilization ranks second, the landfill method ranks third, and the microwave method ranks last.

3.2.2. Evaluating Alternative Medical Waste Disposal Methods by PROMETHEE Method

Visual PROMETHEE application was used for the application of the PROMETHEE method, which is an MCDM method. Visual PROMETHEE Version 1.4.0.0, one of the first real interactive software based on the superiority method, was designed in 2012. It is a highly visual and user-friendly software. The application stages are as follows:

Phase 1: Entering the names and values of alternatives and criteria into the Visual PROMETHEE application.

Phase 2: Entering Min/Max in the ‘Preferences’ section to determine whether the criteria are benefit- or cost-oriented.

Phase 3: Entering the weights of the criteria obtained with CRITIC into the ‘Weight’ section.

Phase 4: To enter the preference functions, the entry is made in the ‘Preference Fn.’ section, but the preference function suggested by the application is selected in determining the preference functions. Data inputs in the application are provided in Figure 8.

Phase 5: Under the ‘Thresholds’ button, the option ‘absolute’ was selected depending on the type of the preference function. The system was then operated. The results are provided in Figure 9.

The ranking was made by using the net priority values of the alternatives with the PROMETHEE method. The graphical representation of this ranking is given in Figure 10.

In conclusion, as a result of ranking by PROMETHEE method, the incineration method ranks first in the selection of medical waste disposal method for the hospital, while steam sterilization ranks second, the landfill method ranks third, and the microwave method ranks last.

3.2.3. Evaluating Alternative Medical Waste Disposal Methods by EDAS Method

Last but not least, alternative medical waste disposal methods were evaluated by the EDAS method. The steps of the EDAS method were calculated individually by using Microsoft Excel 2016.

Firstly, considering the values in the decision matrix created in Table 4, the average solution vector ${AV}_j$ was calculated by the formula in Equation (A24). The resulting average solution matrix is provided in

Table 15. Calculating the average solution matrix.

Decision Matrix
Alternatives/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
	Min	Min	Min	Min	Max	Max	Max	Max	Max	Max	Min	Max
Incineration	9	6	4	9	9	1	7	9	9	9	3	10
Steam Sterilization	6	8	3	5	5	6	4	7	5	1	7	10
Microwave	7	7	3	4	4	3	3	6	3	1	6	9
Storage	3	3	5	3	3	2	6	4	1	1	9	10
Average Solution Vector
AVj	6.25	6	3.75	5.25	5.25	3	5	6.5	4.5	3	6.25	9.75

.

Then, using Equations (A27) and (A29), positive distances from the average solution were calculated as shown in

Table 16. Calculating positive distances from the average solution.

Positive Distance Matrix
Alternatives/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
	Min	Min	Min	Min	max	Max	Max	Max	Max	Max	Min	Max
Incineration	0.000	0.000	0.000	0.000	0.714	0.000	0.400	0.385	1.000	2.000	0.520	0.026
Steam Sterilization	0.040	0.000	0.200	0.048	0.000	1.000	0.000	0.077	0.111	0.000	0.000	0.026
Microwave	0.000	0.000	0.200	0.238	0.000	0.000	0.000	0.000	0.000	0.000	0.040	0.000
Storage	0.520	0.500	0.000	0.429	0.000	0.000	0.200	0.000	0.000	0.000	0.000	0.026

.

As a next step, negative distances from the average solution were calculated using Equations (A28) and (A30) as shown in

Table 17. Calculating negative distances from the average solution.

Negative Distance Matrix
Alternatives/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
	Min	Min	Min	Min	Max	Max	Max	Max	Max	Max	Min	Max
Incineration	0.440	0	0.067	0.714	0	0.667	0	0	0	0	0	0
Steam Sterilization	0	0.333	0	0	0.048	0	0.200	0	0	0.667	0.120	0
Microwave	0.120	0.167	0	0	0.238	0	0.400	0.077	0.333	0.667	0	0.077
Storage	0	0	0.333	0	0.429	0.333	0	0.385	0.778	0.667	0.440	0

.

As a next step, the weight values from the CRITIC method in Table 7 were multiplied by the positive distance matrix in Table 16 to obtain the weighted total positive distance matrix in

Table 18. Calculating weighted total positive distances.

Weighted Positive Distance Matrix
Alternatives/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
	Min	Min	Min	Min	Max	Max	Max	Max	Max	Max	Min	Max	Spi	NSPi
Incineration	0	0	0	0	0.047	0	0.030	0.025	0.064	0.153	0.035	0.002	0.356	1
Steam Sterilization	0.004	0	0.020	0.005	0	0.094	0	0.005	0.007	0	0	0.002	0.138	0.389
Microwave	0	0	0.020	0.027	0	0	0	0	0	0	0.003	0	0.050	0.141
Storage	0.052	0.047	0	0.049	0	0	0.015	0	0	0	0	0.002	0.165	0.464

.

As a next step, the weight values from the CRITIC method in Table 7 were multiplied by the negative distance matrix in Table 17 to obtain the weighted total negative distance matrix in

Table 19. Calculating weighted total negative distances.

Weighted Negative Distance Matrix
Alternatives/Criteria	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
	Min	Min	Min	Min	Max	Max	Max	Max	Max	Max	Min	Max	SNi	NSNi
Incineration	0.044	0	0.007	0.082	0	0.063	0	0	0	0	0	0	0.196	0.212
Steam Sterilization	0	0.031	0	0	0.003	0	0.015	0	0	0.051	0.008	0	0.108	0.565
Microwave	0.012	0.016	0	0	0.016	0	0.030	0.005	0.021	0.051	0	0.006	0.157	0.371
Storage	0	0	0.034	0	0.028	0.031	0	0.025	0.050	0.051	0.030	0	0.249	0

.

Finally, the evaluation scores for each medical waste disposal method were calculated using Equation (A35), resulting in the values in

Table 20. Final ranking of alternative methods according to EDAS method.

	Evaluation Scores	Ranking
Incineration	0.606	1
Steam Sterilization	0.477	2
Microwave	0.256	3
Storage	0.232	4

.

In conclusion, as a result of ranking by the EDAS method, the incineration method ranks first in the selection of medical waste disposal method for the hospital, while steam sterilization ranks second, the microwave method ranks third, and the landfill method ranks last.

3.3. Sensitivity Analysis

MCDM processes enable decision-makers to evaluate multiple and often conflicting criteria simultaneously. However, the outputs of these methods (ranking of alternatives) may be sensitive to small changes in inputs, such as the criteria weights used or the performance of alternatives against the criteria. In addition, there may be personal differences in the evaluations of criterion weights by the decision-makers over time. Sensitivity analysis is performed to measure such sensitivity, to increase the reliability of the results and to make more effective decisions on the effects of changes.

To this end, in this section of the study, 10 different scenarios were identified to test the robustness of the results from TOPSIS, PROMETHEE, and EDAS methods used in the selection of medical waste disposal methods at the university hospital in Adana. These scenarios analyzed all three methods for different criteria weights and sensitivity results are indicated separately. Separate sections indicate how the ranking of the methods varies.

Table 21. Determined scenarios.

	ICC	OC	ER	E	APC	F	C	P	TOW	ERE	WR	PA
Importance Weights	0.101	0.093	0.101	0.115	0.066	0.094	0.074	0.065	0.064	0.077	0.068	0.082
Scenario 1	0.082	0.068	0.077	0.115	0.066	0.094	0.074	0.065	0.064	0.101	0.093	0.101
Scenario 2	0.083	0.083	0.083	0.083	0.083	0.083	0.083	0.083	0.083	0.083	0.083	0.083
Scenario 3	0.082	0.093	0.101	0.115	0.066	0.094	0.074	0.065	0.064	0.077	0.068	0.101
Scenario 4	0.064	0.065	0.066	0.068	0.074	0.077	0.082	0.093	0.094	0.101	0.101	0.115
Scenario 5	0.101	0.093	0.131	0.105	0.066	0.084	0.074	0.055	0.064	0.077	0.068	0.082
Scenario 6	0.000	0.093	0.101	0.011	0.066	0.094	0.179	0.115	0.114	0.077	0.068	0.082
Scenario 7	0.111	0.113	0.111	0.115	0.006	0.114	0.104	0.005	0.004	0.107	0.108	0.102
Scenario 8	0.093	0.085	0.093	0.105	0.061	0.086	0.068	0.060	0.059	0.154	0.062	0.075
Scenario 9	0.093	0.085	0.129	0.105	0.084	0.086	0.068	0.059	0.059	0.070	0.087	0.075
Scenario 10	0.099	0.091	0.099	0.124	0.071	0.092	0.073	0.064	0.063	0.076	0.067	0.081

shows 10 different scenarios (Scenario 1, Scenario 2, Scenario 3, Scenario 4, Scenario 5, Scenario 6, Scenario 7, Scenario 8, Scenario 9, and Scenario 10) which were determined by the CRITIC method and used in the ranking of the alternatives by making changes in the weights for each criterion.

The changes made to the scenarios are as follows:

Scenario 1: Energy recovery, waste residue, and public acceptance have been increased, while emissions, energy recovery, and public acceptance have been maintained at a high level.

Scenario 2: All criteria have been weighted equally.

Scenario3: Emissions, environmental risk, and public acceptance have been prioritized.

Scenario 4: Public acceptance, waste residue, energy recovery, waste diversity, waste reduction volume, capacity, and air pollution control have been increased.

Scenario 5: Environmental risk, emissions, and initial investment cost have been maintained at a high level.

Scenario 6: Operational factors such as capacity, productivity, and waste diversity have been strongly prioritized. The initial investment cost weight has been reduced to zero.

Scenario 7: Air pollution control, productivity, and waste diversity have been significantly reduced.

Scenario 8: The energy recovery criterion has been given the highest weight and doubled.

Scenario 9: The importance values of the environmental risk criterion, air pollution control, and waste residue criteria have been increased.

Scenario 10: The emissions criterion has been given the highest weight.

3.3.1. Sensitivity Analysis for TOPSIS

The change in the ranking of alternative medical waste methods as a result of evaluation by the TOPSIS method for ten different scenarios is indicated in Figure 11.

Following sensitivity analysis for TOPSIS method, it was observed that the change of criteria weights in the scenarios did not lead to a significant change in the ranking of alternative medical waste methods. It was observed that only in Scenario 4, the microwave method and landfill method swapped their ranking spots.

3.3.2. Sensitivity Analysis for PROMETHEE

The change in the ranking of alternative medical waste methods as a result of evaluation by the PROMETHEE method for ten different scenarios is indicated in Figure 12.

Following sensitivity analysis for PROMETHEE method, it was observed that the ranking of alternative medical waste methods changed in Scenario 6 and Scenario 9, following the change of criterion weights in the scenarios. It was also observed that the microwave method and landfill method swapped their ranking spots.

3.3.3. Sensitivity Analysis for EDAS

The change in the ranking of alternative medical waste methods as a result of evaluation by the EDAS method for ten different scenarios is indicated in Figure 13.

Following sensitivity analysis for EDAS method, when the ranking of alternative medical waste methods was examined with the change of criteria weights in the scenarios, it was observed that, only in Scenario 7, the microwave method and landfill method swapped their ranking spots.

When the sensitivity analyses were examined, it was generally observed that the ranking of alternative medical waste methods did not change significantly for all three decision-making methods. As a result of the analysis, it is noted that the results of this study are reliable. In the evaluation of the university hospital in Adana, incineration and steam sterilization methods ranked first and second, respectively. In general, the microwave method ranks third while the sanitary landfill method ranks last.

Overall, the sensitivity analysis graph provides a better understanding of the relative advantages of the four waste treatment options:

Incineration: Its ranking as the first option in most scenarios stems from its advantages of high capacity (C), high energy recovery (ERE) potential, and the ability to process a wide variety of waste types (TOW). However, its ranking may fall behind steam sterilization, particularly in scenarios where the weightings of emissions (E) and environmental risk (ER) criteria change (as in Scenario 10). High initial investment costs (ICC) and air pollution control (APC) requirements may be disadvantageous in situations where cost/environmental weights increase.

Steam Sterilization: Its consistent ranking as the second option in most scenarios stems primarily from its ability to sterilize waste, reducing environmental risk (ER) and emissions (E), its relatively better public acceptance (PA), and its lower operating costs (OC). However, its capacity (C) and energy recovery (ERE) potential may be lower than incineration. The rise to first place in Scenario 10 suggests that this scenario places greater emphasis on environmental/operational performance or criteria other than ICC/OC.

Microwave and Storage: These two options consistently remain in third and fourth places. While storage generally has the lowest initial investment costs (ICC), it suffers from the disadvantages of generating the highest environmental risk (ER), the lowest public acceptance (PA), and the highest residual waste (WR). While microwave offers similar advantages to steam sterilization as a sterilization method, it may be weaker in terms of capacity (C) and productivity (P). The temporary ranking change in Scenario 4 indicates that microwave sterilization can outperform storage in operational or environmental terms in certain scenarios, but this is not consistent.

In summary, incineration and steam sterilization are the two strongest options, potentially replacing each other depending on the criteria’s weighting. Incineration stands out in economic/capacity-focused scenarios, while steam sterilization stands out in environmental/social scenarios. Microwave and storage, however, are ranked lower, largely due to their disadvantages.

This sensitivity analysis helps us understand the reasons for the ranking changes. Considering these changes:

Cost/Operational (ICC, OC, C, P, F, TOW): When operational criteria (capacity, productivity) are given a higher weight, as in Scenario 6, incineration remains superior to steam sterilization. This is because incineration generally offers a higher capacity and a wider range of waste disposal options.

Environmental (E, ER, APC, WR): In Scenario 10, the weighting of the emission (E) criterion is significantly increased. In this case, it is logical that steam sterilization (environmentally friendly sterilization) would surpass incineration (higher emission potential) and move to first place. Although environmental factors are similarly high in Scenario 9, the ranking was maintained because the emission (E) criterion was not as dominant as in Scenario 10.

In terms of energy (ERE): the highest weighting given to energy recovery (ERE) in Scenario 8 strengthened the top-ranking position of incineration, which has a high ERE potential.

These detailed weightings clearly demonstrate which criteria increase makes each disposal method more attractive.

3.4. Possible Political Implications of the Findings

The conclusions of this study regarding incineration could have profound and multifaceted implications for waste management policies in Adana and similar metropolitan cities. For example:

• From an infrastructure and investment perspective:

Trend Towards Incineration Facilities: If this conclusion is accepted by local governments, large-scale infrastructure investments can be focused on establishing a centralized incineration facility (waste-to-energy facility with energy recovery) rather than promoting other non-incineration technologies (e.g., autoclaves) in the short term.

Changing Cost Structure: The high initial investment costs of incineration facilities require public–private partnerships or long-term municipal borrowing. This may lead to municipalities reframing their financial priorities and increasing waste tariffs.

• From an environmental management and public health perspective:

Air Quality Monitoring: Incineration inevitably creates air emissions (dioxins, furans, heavy metals). Politicians will be forced to acknowledge the importance of environmental and public health criteria to support this decision. This necessitates the tightening of existing environmental regulations and the continuous, transparent, and independent monitoring of the facility’s environmental impact.

Energy Recovery: The greatest policy benefit of an incineration facility is the generation of electricity or heat through waste-to-energy (WTE). This integration with local energy policies ensures that the facility’s environmental costs (emissions) are offset by renewable energy production.

• Regarding Legal Framework and Technology Incompatibility:

Regulatory Alignment: If existing local legislation focuses more on non-incineration technologies (autoclaving), this study provides a scientific basis for reassessing existing medical waste disposal regulations in terms of comprehensive waste volume and types.

Public Perception Management: The public generally has a negative perception of incinerators (due to odor and pollution concerns). This policy decision will require a comprehensive and transparent communication strategy (risk communication) regarding facility location selection and public information (social criteria).

4. Conclusions and Recommendations

In Turkey, studies on medical waste management have generally focused on analyzing the overall situation at a regional or national level, identifying existing problems and proposing solutions.

This study ranked the methods used for the disposal of medical wastes at the university hospital in Adana province. The rankings by all methods are provided in

Table 22. Ranking of alternative medical waste disposal methods according to TOPSIS, PROMETHEE, and EDAS methods.

	TOPSIS Method	PROMETHEE Method	EDAS Method
Incineration	1	1	1
Steam Sterilization	2	2	2
Microwave	4	4	3
Storage	3	3	4

and Figure 14.

Figure 14 shows that TOPSIS and PROMETHEE methods provided the same results in ranking. The incineration method ranked first, while steam sterilization ranked second. In EDAS method, the microwave irradiation and sanitary landfill methods swapped places in rankings.

In addition, Spearman Rank Correlation Analysis was applied to determine the relationship between the ranking results obtained from different methods. This test checks the statistical significance of the results obtained from the applied model. The results of the rank correlation test for all applied methods are shown in

Table 23. Spearman correlation test results between methods.

	TOPSIS	PROMETHEE	EDAS
TOPSIS	1.00	1.00	0.80
PROMETHEE	1.00	1.00	0.80
EDAS	0.80	0.80	1.00

.

A correlation of 1.00 between TOPSIS and PROMETHEE indicates that these two methods rank the alternatives identically. The EDAS method exhibits a very high positive correlation of 0.80 with both TOPSIS and PROMETHEE. This indicates strong agreement in the overall ranking trend of the methods, but little variation in the rankings of all alternatives. Consequently, the results of the validation test indicate that the proposed rankings are validated and can be considered reliable.

In line with all the criteria determined as a result of the analyses and expert opinions, the most appropriate medical waste disposal methods for the university hospital in Adana province were clearly ranked. This ranking is an objective reflection of current local conditions and strategic priorities. The resulting sensitivity analyses confirmed the robustness of the selected methods by showing how small changes in the decision parameters (change in criterion weight) affect the final ranking.

4.1. Reasons of Method Selection

In this study, the integration of four different Multi-Criteria Decision-Making (MCDM) methods is a methodological necessity in addressing a complex and multi-dimensional decision problem such as the selection of a medical waste disposal method. The main rationale for this integration is to combine the unique strengths of each method and eliminate potential weaknesses and biases that might arise from considering a single method separately. Firstly, the use of the CRITIC method in determining the criterion weights added a high degree of objectivity to the study process. When determining the criteria weights, CRITIC statistically analyzes the contrast intensity (standard deviation) and correlation (discrepancy) between the criteria, rather than the subjective assessments of the decision-makers [5,26]. In this way, CRITIC provides a scientific basis that reflects the true degree of discrimination of the criteria in the context of the problem. The objective determination of the criterion weights reduces the potential weaknesses of subjective weighting methods (such as AHP) and the susceptibility of the results to bias [5].

The combination of CRITIC–TOPSIS–PROMETHEE–EDAS offers significant added value and increased reliability compared to considering each method separately. This added value stems from the fact that the three ranking methods used (TOPSIS, PROMETHEE, and EDAS) validate and complement each other based on different mathematical principles: TOPSIS is based on the principle that the alternative is closest to the ideal solution and farthest from the negative ideal solution [28]. EDAS, on the other hand, uses a completely different logic that measures the positive and negative distance of the alternative from the average solution [30]. This different approach of EDAS produces results that are more robust to potential problems, such as ranking reversals, that can theoretically occur in ideal solution-based methods like TOPSIS [16]. A third method, PROMETHEE, confirms the findings of other distance-based methods from a third perspective based on the superiority principle by ranking them based on pairwise comparisons and net superiority flows [29]. Consequently, combining these methods maximizes the robustness of the decision and systematically and reliably proves that the chosen burn-in method is the most suitable alternative, without remaining within the limits of a single algorithm.

4.2. Original Contributions of the Study

The Adana study offers significant and original contributions to the international MCDM literature on medical waste disposal method selection in three key areas:

• Innovative and proven hybrid methodology: The study presents a robust hybrid framework that combines the CRITIC method for criteria weighting and three different ranking methods: TOPSIS, PROMETHEE, and EDAS for ranking alternatives. While most of the literature [17,31] relies on subjective opinions of experts for weighting, this study used CRITIC to objectively derive weights based on intercriterion correlation and information content. This reduces the risk of the results being influenced by personal biases or subjective judgments. The simultaneous validation of the incineration result by three different mathematical principle-based methods—TOPSIS (closeness to ideal solution), PROMETHEE (predominance flow), and EDAS (deviation from mean distance)—is a significant novelty, demonstrating the robust methodological stability and reliability of the decision. This approach minimizes the risk of potential ranking inconsistencies among studies ranking using only one method (e.g., PF-SWARA-ARAS, DE-MATEL-MULTIMOORA).
• Conclusion contradictory to literature and emphasizing local conditions: The study’s conclusion that incineration is the most suitable alternative in Adana contradicts the dominant trend in the international and national literature (autoclaving/steam sterilization). This contradiction is one of the study’s major theoretical contributions. The results demonstrate that criteria such as absolute technical capacity, volume reduction efficiency, and comprehensive waste disposal capabilities favor incineration technology (e.g., for chemical and pharmaceutical wastes where non-incineration technologies such as autoclaves and microwaves are inadequate), despite environmental and cost concerns. This finding demonstrates that the optimal solution for medical waste disposal is not universal, but rather depends on the local and operational constraints of a large metropolitan area (Adana) or a university hospital with a complex waste stream (e.g., the need for single-center disposal of high-volume hazardous waste).
• Perspective on Turkish studies literature: Studies on medical waste disposal method selection in Turkey have generally favored autoclave/steam sterilization in regions such as Erzurum [7] or Sivas [11]. This study is among the first to show how the technical advantages of incineration, combined with the objectivity of CRITIC, became the primary decision driver in a large metropolitan area close to the Mediterranean region.

4.3. Limitations of the Study

Despite its methodological strength and original findings, this study has some limitations, both common and unique to other studies in the literature:

• Methodological limitations and data constraints: While the CRITIC method objectively determines criterion weights, the evaluation of alternatives against criteria (decision matrices used in the TOPSIS, PROMETHEE, and EDAS methods) is based entirely on expert opinions. There are several important reasons for this:

  ✓

  Institutional information such as operational data, costs, waste processing performance, sterilization efficiency, or emission measurements in healthcare institutions in Turkey is not open to external sharing due to legal, ethical, and institutional confidentiality reasons.

  ✓

  Due to relevant regulations, the hospital management did not approve of reporting the numerical data requested within the scope of the study. Therefore, the study had to rely solely on the knowledge, field experience, and professional judgment of experts.

  ✓

  Expert assessments are also widely used in the international literature (such as Demir & Moslem 2024 [11]), where similar data limitations exist.

Therefore, although the study follows an approach accepted in the literature, the results are limited to the opinions of a specific group of experts. The number of experts, their level of experience, professional background, and evaluation methods lead to a certain level of subjectivity in the results. However, all of the experts’ evaluations are presented in detail in the Appendix B, ensuring transparency and traceability.

• Data uncertainty and lack of fuzzy modeling: A second important limitation of the study is that the methods used are MCDM methods that work with precise values. This leads to the following conclusions:

  ✓

  Hesitation, indecision, uncertainty, or evaluation intervals in the experts’ evaluations are not directly represented in the structure of the methods.

  ✓

  Experts’ uncertainty about some criteria cannot be numerically reflected in the scoring.

  ✓

  Medical waste management is an inherently risky, uncertain, multi-stakeholder, and multivariate field. Therefore, uncertainty modeling is crucial for realistic decision support.

Although the study indirectly tested uncertainty by conducting a comprehensive sensitivity analysis across more than 10 scenarios, this approach does not mathematically model the sources of uncertainty and only measures the system’s stability against changes in the criteria weights. However, methods used in the literature, such as intuitionistic hesitant fuzzy sets (IHF), fermatean fuzzy sets, and intuitionistic fuzzy sets, can more effectively represent the uncertainty and binary uncertainty in expert assessments [12,14]. Therefore, modeling uncertainty with more advanced methods is an important area for future research.

• Application and contextual limitations and limited variety of alternatives: The study evaluated only four disposal methods. This limitation is not a direct methodological choice; it stems from regulatory, infrastructure, and institutional capacity constraints in the application area:

  ✓

  Only certain disposal methods are technically applicable in the university hospital under study.

  ✓

  Current legislation in Turkey limits the use of medical waste disposal methods.

  ✓

  Many methods, such as plasma pyrolysis, certain types of chemical oxidation, or advanced microwave systems, were excluded due to their infeasibility both institutionally and in terms of technical infrastructure.

• Therefore, although the number of alternatives appears small, the selected methods represent all applicable options for the hospital. However,

  ✓

  The limited diversity of alternatives hinders the comparison of new technologies or different regional solutions.

  ✓

  The study’s focus on a single university hospital in Adana limits the generalizability of the results.

  ✓

  As emphasized by Anjum et al. (2024), waste types, waste volume, personnel structure, and management strategies can vary considerably across different healthcare institutions [22].

Therefore, the findings of the study may not be directly applicable to healthcare institutions of different sizes.

• Inability to dynamically model socioeconomic and environmental externalities: Although environmental criteria were evaluated within the scope of the study, the evolution of externalities over time was not examined comprehensively enough:

  ✓

  Incineration, in particular, involves high emissions, high investment costs, and long-term operating expenses. The long-term sustainability of these costs may vary depending on economic fluctuations and energy costs.

  ✓

  Elements such as public health, public perception, and social acceptance can vary over time, and the current model does not dynamically represent these components.

Although Lee et al. (2016) emphasized the dominance of cost, the high initial investment and long-term operating costs of incineration may threaten the long-term sustainability of the decision in this study [31].

4.4. Future Research Directions

Given the limitations of this study and the gaps in the literature, future research could focus on the following areas:

• Advanced fuzzy integration: CRITIC and PROMETHEE methods could be integrated with interval-valued, Pythagorean, or T-Spherical Fuzzy sets (e.g., as in Anafi et al. [15]). This would increase methodological robustness by combining objective criterion weighting (CRITIC) with uncertainty modeling (Fuzzy).
• Fuzzification of compromise methods: The advantages of incineration could be retested under uncertain conditions using fuzzy versions of PROMETHEE and EDAS.
• Dynamic modeling of decision criteria: DEMATEL or ANP (Analytical Network Processing) could be used in conjunction with CRITIC to model the interdependencies between criteria. As highlighted in Liu et al. (2015), feedback between criteria plays a critical role in medical waste disposal selection [3].
• Expansion of social criteria: For the sub-criteria of the social dimension highlighted by Demir and Moslem (2024) (social acceptance, risk perception, occupational safety), more precise scales, perhaps supported by geospatial data (GIS), could be developed [11].
• Multi-region scope and life cycle assessment: The incineration outcome in Adana could be retested in other major Turkish metropolitan areas (İzmir, Ankara) with different demographic and waste profiles, thus further generalizing the findings.
• Life cycle analysis (LCA): Instead of just the immediate cost and environmental impact, the environmental and cost impacts of incineration and autoclaving alternatives throughout their entire life cycle (from installation to final disposal) could be integrated into the MCDM model.

In conclusion, the MCDM methods have proven to be an indispensable approach in bringing solutions to vital and multi-dimensional problems such as medical waste disposal.