Study of landfill site suitability and sensitivity analysis using GIS-FAHP for solid waste management: a case study in Ranchi, India

1 Introduction

The generation of municipal solid waste (MSW) is an inherent characteristic of nations that experience urbanisation, industrialisation, and urban sprawl. Consumerism and changing lifestyles have compounded the generation of non-recyclable waste such as single-use plastic (SUP), disposable products and packaging waste, all forming a part of throwaway products (OECD, 2022). If this MSW is mishandled, it leads to unsanitary disposal in low-lying areas, beside roads, directly in the streams, or by open burning, which adversely impacts air, soil, surface water, groundwater, and biodiversity (Ali and Ahmad, 2020).

India is also undergoing rapid urbanisation owing to growth in Gross Domestic Product (GDP), boom in the IT sector, population growth, and industrialisation. This has resulted in the generation of more than 62 million tonnes of waste in 2021–22, which includes plastic, paper, cloth, food waste, etc (CPCB, 2022) from households, educational institutions, and commercial offices, among other sources. Handling of MSW in India falls under the purview of Urban Local Bodies (ULBs) and panchayats, guided by the Solid Waste Management Rules, 2016, Central Pollution Control Board (CPCB) guidelines, and Municipal Solid Waste Management Manual by Central Public Health and Environmental Engineering Organisation (CPHEEO) (Kapilan and Elangovan, 2018). Statistically, MSW collection efficiency is nearly 91%, of which 24% of the collected waste is disposed of in sanitary landfills, whereas 22% of the total waste generated is neither collected, processed, nor disposed (CPCB, 2022). Therefore, emphasis should be placed on effective solid waste management (SWM) to process MSW effectively and mitigate its harmful impacts on the environment, biodiversity, and human health.

SWM consists of waste generation, segregation, primary and secondary collection, storage, transfer and transport, waste processing, and disposal, emphasising resource efficiency, resource conservation, and reducing waste disposal (MOUD, 2016). To ensure environmentally friendly handling, waste management is prioritised in a tiered hierarchy focusing on prevention, reduction of generation, reuse, recycling, resource recovery, co-processing, incineration, and landfilling, from most preferred to least preferred. Based on the waste management hierarchy, landfilling is the least preferred option, yet it remains the most widely used globally. In India, most waste is landfilled owing to the nascent market of MSW and the added advantage of cost-effectiveness, affordability, and straightforward application (MOUD, 2016). This puts pressure on city planners and local bodies with scarce land to identify waste disposal sites and transport waste to them, while simultaneously complying with the criteria laid down in national statutes and guidelines.

Conventionally, landfill identification is performed by physically shortlisting suitable sites based on haulage time and land availability (Yesilnacar et al., 2012). For cost-effective and scientific identification of landfill sites, Geographic Information System (GIS) and Remote Sensing (RS) have become important tools globally, which can manage vast amounts of geographic information, such as location coordinates, slope, aspect, LULC, etc. (Yesilnacar et al., 2012). RS is a technique that remotely captures data associated with a location, whereas GIS is a technology that captures, stores, analyses, and displays this geospatial information (Chang, 2019). It serves the objective of performing a primary screening and identifying suitable sites in accordance with statutory and scientific criteria. Recently, RS and GIS have become crucial techniques for gathering and assimilating geographical information for land-use feasibility studies such as rainwater harvesting systems, solar power plant and waste landfill site identification, and identification of sources of heavy metal contamination in soil, among others (Mahapatra et al., 2020; Rane et al., 2024; Ünsal et al., 2024; Zhong et al., 2022).

Identification of suitable landfill sites involves evaluating multiple criteria, such as LULC, population, slope, soil, proximity to rivers, roads, and lakes (Singh et al., 2023). Hence, the multi-criteria decision-making (MCDM) procedure is being widely used with GIS to optimize decision-making between environmental, human, geological, hydrological, and land use criteria (Ali and Ahmad, 2020; Sharifi et al., 2009). Economically and environmentally suitable, and socially acceptable landfill sites have been identified using various MCDM techniques such as Analytical Hierarchy Process (AHP) and Fuzzy AHP (FAHP) (Khan and Samadder, 2015).

AHP, introduced by Saaty in 1980, is an MCDM tool that works on four major steps- (1) breaking the problem into small constituents and structuring them in hierarchical form; (2) making pairwise comparisons among these constituents; (3) estimating relative weights; and (4) aggregating relative weights and synthesising to arrive at a decision (Saaty, 1987). It is a helpful tool because it addresses both qualitative and quantitative aspects of decision-making and organises them in a hierarchical structure. Studies have been conducted worldwide that have employed the GIS and AHP frameworks to identify potential landfill sites in their cities or municipalities. Majid and Mir (2021) identified 5 potential landfills in Srinagar, India, sites by combining GIS with AHP. Malo et al. (2024) also applied GIS- AHP framework to identify landfill sites in Balurgarh, India and located 10 potential landfill sites. Another study by İskender Soyaslan (2025) in Burdur Lake Basin, Turkiye, for landfill site suitability, found that 22.38% of the study area was highly suitable for development as a landfill. In these studies that have used AHP as an MCDM tool, a comparison matrix was created using crisp numbers derived from expert judgement, literature or author concurrence. The comparison matrix created on crisp numbers relies more on human judgement, handles less uncertainty and increases bias and subjectivity. In such cases, uncertainty in expert judgement can yield imprecise results, and a small change in input value can sometimes change ranking significantly (Chen et al., 2025; Tripathi et al., 2022).

To overcome the shortcomings of AHP due to uncertainty in human preferences, fuzzy sets have been incorporated into the AHP process to represent imprecise data. FAHP uses fuzzy comparison ratios in various methods, such as the fuzzy least square method (Xu, 2000), geometric mean method (Buckley, 1985), and synthetic extend analysis (D.-Y. Chang, 1996). In recent times, FAHP is becoming a popular MCDM method in land use suitability studies for siting landfills, solar and wind farms, rainwater harvesting systems, among others (Ahmad et al., 2024; Ali and Ahmad, 2020; Mahapatra et al., 2020; Rane et al., 2024; Ünsal et al., 2024; Zhong et al., 2022). Islam et al. (2020)have used Chang’s extent analysis and weighted overlay analysis and identified 15 suitable landfill sites in Rajshahi, Bangladesh. In another study, (Singh et al., 2023), employed FAHP (Buckley’s geometric mean method) with DRASTIC vulnerability-based multi-criteria decision analysis to identify potential landfill sites in Dhanbad, Jharkhand. The Fuzzy AHP method has been widely preferred and applied over AHP, as it is difficult for experts to assign an exact ratio while comparing two criteria (Buckley, 1985) and so FAHP offers a better relationship between multiple criteria.

Some recent studies have also calculated criteria weights from AHP and FAHP methods of MCDM and compared the results to understand their performance in site suitability analysis and groundwater potential analysis (Bastola et al., 2024; Chen et al., 2025; Tripathi et al., 2022) A study by Bastola et al. (2024) found that FAHP slightly outperformed AHP in assessing the current dam locations. Tripathi et al. (2022) studied a site selection for a hospital using AHP and FAHP methods and compared their results in terms of sensitivity analysis in site ranking. The authors observed that FAHP showed only one variation in site ranking, whereas AHP showed constant variation when the criteria weights were varied. In another study by Chen et al. (2025) it was reported that FAHP outperformed AHP, with the Area under the Curve (AUC) of the Receiver Operating Characteristic (ROC) curve for FAHP being slightly higher than that for AHP. Therefore, FAHP is widely preferred over AHP by researchers due to its robustness, reduced sensitivity to changes in criterion weights, and better ROC AUC score.

As MCDM-derived rankings can be conditional, uncertainty may arise from various sources, including data processing, input data, selected criteria, and the associated thresholds. This may arise from decision-makers’ uncertainty about their preferences for the criteria, or from the unknown scale and nature of the criteria. Therefore, sensitivity analysis is employed, which enables the study of the relationship between input parameters and output. It has been identified as a crucial tool for model calibration and validation.

In India, Ranchi, the capital of Jharkhand, urbanisation and changing lifestyles have increased solid waste generation, which is being landfilled at the Jhiri dumpsite. The dumpsite is spread over 42.05 ha of land, receives more than 500 tons of waste daily, and lies within a 20 km radius of the railway station (Tiwary, 2023). As per the Times of India report of April 2023, the landfill has been the cause of health hazards and environmental pollution for people residing in proximity (Tiwary, 2023) and hence, identification of an alternate, scientific landfill site becomes pertinent.

Two studies have been carried out to identify landfill sites in Ranchi, India. The first study by Bera et al. (2018) employed GIS and derived weights for 9 criteria by analysing major and minor interrelationships between two factors, assigning a weight of 1.0 and 0.5 to the effect of major and minor factors, respectively. However, the study has applied fixed weights to major and minor factor effects, has not conducted pairwise comparisons among the criteria, and lacks a consistency check. Also, the study has not included an important regulatory criterion, as per the SWM Rules, 2016, which is the distance from the airport as a criterion for locating landfill sites (MoEFCC, 2016). In another study carried out by (Shubham and Singh, 2025), GIS was used with AHP to site landfill sites for Ranchi, by carrying out a pairwise comparison of 9 criteria. The study found the groundwater table to have the highest criteria weight and 12.38% area as most suitable for the site of a landfill. But the study has used AHP, which inherently biases towards experts’ judgement, has not accounted for the minimum area of landfill by estimating population growth, nor has it considered the distance of the landfill from the city.

In this respect, the present study is necessary and pertinent to delineate potential landfill sites by combining GIS with MCDM (using FAHP) and incorporating essential criteria, such as LULC, distance from the airport, slope, among others, that influence landfill site identification in Ranchi. Previous studies have incorporated GIS to display spatial data and relationships, and MCDM such as AHP and FAHP to assign criteria weights to identify landfill sites, this study provides a holistic approach towards sustainable waste management by employing a GIS- FAHP framework and incorporating aspects of sustainability such as measuring the landfill capacity to accommodate future waste based on population and cost efficiency by reducing transportation cost, along with carrying out a sensitivity analysis. Based on the author’s knowledge, only a limited number of studies in India have calculated the minimum landfill area and conducted a sensitivity analysis by varying the criterion weights. The study is holistic in its approach to sustainable landfill site identification as it aims

• To identify potential landfill sites in Ranchi, India by using the GIS- MCDM framework.
• To estimate the population growth, quantity of waste generated and minimum area of landfill for a landfill operational time of 20 years.
• To carry out sensitivity analysis to determine the robustness and reliability of model results.

Hence, this research aims to address the following research question-to design a systematic, holistic and scientific GIS- FAHP framework to site a landfill by integrating environmental, social and technical considerations for long-term urban sustainability, and to identify the most critical criteria that exert the greatest influence on the site suitability outcome.

2 Methodology

2.1 Study area

Ranchi is a district in Jharkhand, India, located from 23°21′N to 23°87′N latitudes and 85°75′E to 85°87′E longitudes, with a total area of 5,097 km². It is situated in the southern part of the Chota Nagpur Plateau and has a hilly topography. Nearly 23% of the total district’s area is covered by forests, with a rich diversity of flora and fauna and is drained by rivers such as Subarnarekha, South Koel, Sankh, and Damodar. The district also has reserves of natural resources such as limestone and China clay, which have made it an industrial hub with 6 industrial areas. For administrative purposes, the district is divided into 2 subdivisions- Ranchi and Bundu subdivisions-with 18 Community Development Blocks and 303 panchayats (Census of India, 2011).

Ranchi city, the capital of Jharkhand and administrative headquarters of Ranchi district, is governed by the Ranchi Municipal Corporation (RMC). RMC was created on 15th September 1979 by merging the then Ranchi Municipality, Doranda Municipality, and the Ranchi Doranda Joint Water Board, and now covers an area of almost 175 km². As per the 2011 census, the population of Ranchi district and RMC was 2.9 million and 1 million, respectively, with a decadal population growth rate of the district of 23.98% between the 2001 and 2011 census, and a population density of 572 per sq. km.

As the state’s capital, the city is experiencing high population and economic growth, leading to a large volume of MSW. Figure 1 depicts the location of Ranchi district, the study area of this work, along with the currently operational dumpsite located at Jhiri..

Presently, the waste from Ranchi city is being dumped at Jhiri dumping yard, which has an area of 42.05 ha, and has been operational for more than 2 decades (Tiwary, 2023). Geographically, the site is located at Latitude: 23° 24′41.10″N and Longitude: 85° 15′21.28″E and is well connected by road and rail. The site receives more than 400 tons of fresh waste per day but has more than 18 lakh tons of legacy waste on the premises. Efforts are being made to reduce the legacy waste, but the mixed composition of waste, unsegregated wet waste and dry waste, the slow pace of work due to seasonal fluctuations, and inadequate equipment have rendered the clearing process slow (Yadav, 2026).

The dumping yard is devoid of a proper lining system and methane capture, which has caused soil, groundwater and air pollution, as reported by multiple newspaper reports. Meanwhile, the fresh waste received at the site has also been reported to be mixed–wet waste and dry waste, and the wet fraction is being handled by setting up a Compressed Bio Gas (CBG) plant with 150 TPD of bio-degradable organic MSW as input for conversion into 5,000 kg per day of compressed gas and 25 tons of fermented bio manure, in an area of 8 ha. The biogas plant is also underutilised currently, as RMC has failed to provide the designated quantity of segregated wet waste (The Times of India, 2025).

All these problems have necessitated the creation of a suitable landfill site with proper lining, leachate collection and methane capture system for Ranchi city to handle solid waste.

2.2 Method

The main goal of this study was to develop a replicable, holistic framework for identifying potential MSW landfill sites by creating a scientific GIS-MCDM framework that incorporated expected population growth and associated waste generation in the city for a landfill with an operational life of 20 years. The GIS–MCDM framework was created by identifying suitable criteria, digitising and creating maps, and deriving criterion weights using the FAHP (Buckley’s geometric mean method). Further, a sensitivity analysis was conducted to assess the model’s output robustness by varying the criterion weights.

To address the research objectives, a systematic methodological framework was applied, and the study was carried out in 6 phases, as depicted in Figure 2 and described in detail in the subsequent sections.

2.2.1 Identification of criteria and digitization of maps

Identification of relevant hydrological, topographic, and infrastructural criteria that affect site suitability is the first stage in identifying potential landfill sites. For this study, the criteria were selected based on a survey of literature, relevance of criteria in the study area, and the national statutes developed in this regard such as SWM Rules, 2016. 11 criteria were identified after such consideration which includes LULC, slope, distance from groundwater, distance from a water body, distance from the river, distance from settlement, distance from lineament, distance from road, distance from railway, distance from airport, and soil type.

Next, the maps of these 11 criteria were created, digitised or geo-referenced from multiple sources. The data for the study were obtained from primary and secondary sources and further processed to ensure uniformity in the projection system, etc. The maps of roads, railways, and water bodies, such as lakes, rivers, and airports, were created by tracing features from topographical maps of India and Diva GIS website and updated using Google Earth Pro. A Digital Elevation Model (DEM) for the study area was obtained from Bhuvan: Indian Geo-Platform of ISRO and used to extract slope and lineament data. The LULC data were also obtained after processing Bhuvan data (National Remote Sensing Centre, 2024), soil from the Food and Agriculture Organisation (Food and Agriculture Organisation, 2023), population from the Indian Census data (Census of India, 2011), and water table from the Central Groundwater Board (Ministry of Jal Shakti, 2023, 2024), respectively. Topographical maps were obtained from the Survey of India (Survey of India, 2022). Data obtained from all the sources was updated with Google Earth Pro and OpenStreetMap (OSM). Finally, the maps were geo-referenced to the desired projected coordinate system. ArcGIS software was used in the study to carry out the GIS analysis with the Projected Coordinate System- WGS 1984 UTM Zone 45N.

2.2.2 Buckley’s geometric mean method for deriving weights using FAHP

After identifying the criteria, the next step was to assign weights to them, as each criterion has a different impact on site selection. Weights for each criterion were calculated using various methods of FAHP, such as Buckley’s geometric mean method, Chang’s synthetic extent analysis method, and Vaan Laarhoven and Pedrycz’s FAHP method. Among the different approaches to FAHP, Buckley’s geometric mean method was adopted because it uses fuzzy geometric mean, which preserves the reciprocal property of pairwise comparison matrices derived from AHP. It is relatively easy to compute while providing a strong mathematical foundation, making its use replicable and more popular. Another popular FAHP method is Chang’s Extent Analysis method, which has been criticised for producing irrational priorities and assigning zero weights to important criteria (Gul, 2020; Vinogradova-Zinkevič et al., 2021). This study employed FAHP as per Buckley’s geometric mean method and Fuzzy triangular scales (Singh et al., 2023), as given in Table 1.

TABLE 1. Linguistic scale and fuzzy triangular scale.

In Buckley’s geometric mean FAHP method, a pairwise comparison was carried out between each criterion to judge their relative importance, and hence a pairwise comparison matrix was generated as given in Table 2 (Saaty, 1987). To check the consistency of the scores assigned in the pairwise comparison matrix, the consistency ratio (CR) was calculated using crisp values of AHP. According to Saaty (1980), inconsistency up to 10% is tolerable, implying that CR should be less than 10%. CR and consistency index (CI) are calculated as given in Equations 1, 2, respectively. After ensuring the consistency of the comparison matrix, fuzziness was introduced to calculate the criterion weights.

C R = C I R I (1)

TABLE 2. Pairwise comparison matrix.

LULC, land use land cover.

SL, slope.

DGW, distance from groundwater.

DWB, distance from waterbody.

DR, distance from river.

DS, distance from settlement.

LM, distance from lineament.

DRO, distance from road.

DRA, distance from railway.

DAP, distance from airport.

ST, soil type.

Where CI is the consistency index, and RI is the random index.

C I = λ max − 1 n − 1 (2)

Where λ_max is the largest or principal eigenvalue of the comparison matrix.

Fuzzy numbers were introduced in the pairwise comparison because it isn’t possible to assign exact numbers when comparing two alternatives, thereby reducing ambiguity and vagueness in decision-making. The pairwise comparison was carried out by using a scale ranging from 1 to 9, with 1 for ‘Equally Important’ and 9 for ‘Absolutely Important’. Table 2 lists the pairwise comparison matrix for each criterion, created using the Fuzzy Triangular Scale. The scores have been assigned to each criterion based on the authors’ mutual discussion following comparison with the literature.

The mathematical formula for constructing a pairwise comparison matrix in FAHP is given as per Equation 3 where‾A is a k*k fuzzy positive reciprocal matrix of all paired comparisons for the issues A₁, A₂, … , A_k. The elements in‾A are‾a_ij, where‾a_ji = (a_ij)⁻¹ and‾a_ii = (1/1, 1/1) for all i, described in Equation 3.where α, β, γ, δ are the fuzzy numbers used in the paired comparison, described by (α/β, γ/δ) where 0< α≤ β≤ γ≤ δ.

A ¯ = a ¯ i j , w h e r e a ¯ i , j = α i j β i j , γ i j δ i j (3)

Next, for each criterion, the geometric mean of the fuzzy comparison value of each membership function was calculated as given in Equation 4. Then, for each triangular membership function of each criterion, the vector summation was carried out, and the inverse of the summed vector was calculated. The weights were then fuzzified using Equation 5.

p i l , m , u = ∏ j = 1 k a i j ¯ 1 k (4)

Where p_i(l,m,u) is the geometric mean of lower (L), middle (m), and upper (u) members of the fuzzy triangular scale of each row.

k is the total number of criteria

w i ¯ = p i × p 1 + p 2 + … + p k − 1 w i ¯ = l w i , m w i , u w i (5)

Where‾w_i is the weight for each positive reciprocal matrix‾A.

The final weights were obtained after normalising, thereby removing units, to improve data integrity and reduce redundancy. The normalised weights of each criterion were used to overlay layers and derive a suitability map.

2.2.3 Creation of restricted area and landfill suitability map

Restricted areas were created around the criteria to limit the detrimental impacts of landfill on the surrounding environment, human health and biodiversity. This exclusion of environmentally sensitive and socially unacceptable locations was carried out in accordance with statutory requirements, such as the SWM Rules 2016 and the literature. The restriction can be due to the minimum distance from a feature, such as the unsuitability of land within 20 km of an airport, or to the characteristics of the land, such as the prohibition of landfill sites on water bodies or built-up land.

To accommodate these restrictions, restricted areas were created around the feature using geoprocessing tools, demarcating the restricted zones from sites with the potential to be developed as landfills. For 9 of the 11 criteria, the Euclidean distance tool in ArcGIS was used to create 5 distance classes for each criterion, and a rating was assigned to each class. For the remaining 2 criteria- LULC and soil type, a rating was provided as per the nature and feasibility of landfill siting. Table 3 lists the 11 criteria and the classes created under each criterion, and Figure 3 depicts the associated maps.

TABLE 3. Reclassified Parameter and their Class Rating.

After the restricted classes for each criterion were created and their relative weights assigned, the maps were overlaid with their relative weights using the Weighted Overlay tool in ArcGIS to produce landfill suitability classes. The landfill suitability class demarcated the entire study area into unsuitable, less suitable, moderately suitable and most suitable for landfill siting.

The landfill suitability map was reevaluated further by selecting locations based on landfill operational time consideration, distance from key infrastructures such as airport, and other socioeconomic factors. The next step was to calculate the minimum area required for the desired landfill site.

2.2.4 Estimation of landfill area for 20 years

As per Schedule 1 of the SWM Rules, 2016, landfills should be large enough to receive waste for at least 20 years. So, estimating the area of land required to receive vast quantities of MSW becomes pertinent. To do so, the study has estimated the population of Ranchi for a 20-year period, between 2024 and 2044, and calculated the total waste generated during this period, and finally estimated the area of land required to landfill this waste.

To do so, population growth was first calculated using the geometric increase method (Equation 6), assuming a constant population growth rate that compounded every decade relative to the existing population (Santosh Kumar Garg, 1977). Census data between 1981 and 2011 were used to calculate the assumed population growth rate r ‾, and it was calculated in 2 steps- (1) calculating the growth rates of population in the 3 decades- 1981–1991, 1991–2001, and 2001–2011; (2) calculating the arithmetic mean of the 3 growth rates as given in Equation 6. Arithmetic mean of the population growth rate was used because it is higher than geometric mean and hence provides a conservative larger value of estimated population (Garg, 1977). Using this growth rate, the population of Ranchi was forecasted for 2024–2044 using the geometric population growth method.

P k = P 0 · 1 + r ¯ k (6)

Where P(k) is the future population after k decades.

P₀ is the initial population, that is, the population at the end of the last known census.

‾r is the assumed growth rate given in Equation 7,

r ¯ = r 1 + r 2 + ⋯ + r k k (7)

Where r₁, r₂, r_3, r_k are the decadal population growth rates for decades 1,2, 3, … , k.

Further, the total waste generated in the 20 years was calculated. As per Swachh Bharat Mission 2.0, for urban local bodies (ULB) with a population greater than 10 lakhs, the per capita MSW generation was 0.55 kg/day. This was taken as the average per capita waste generation rate for the entire 20-year period. Total waste generated for the period was then calculated as per Equation 8.

W = ∑ i = 1 n p i · 0.55 (8)

Where W is the total waste generated in 20 years

p(i) is the population in the ith year

n is the total number of years, which is 20 years for this study.

0.55 is the constant per capita MSW generation for Ranchi

Next, to estimate the minimum area of landfill required, the landfilling rate for Ranchi was taken as 0.49, which was the state average landfilling rate for Jharkhand as per the Annual Report of 2020–21 on Implementation of Solid Waste Management Rules, 2016. Further, the incoming waste to the landfill was assumed to be compacted to a density of 850 kg/m³ before reaching the landfill (Memon and Agrawal, 2019). Potential landfill sites with an area less than the minimum estimated landfill area were removed to ensure landfill longevity.

2.2.5 Identification of potential sites

Refinement of the most suitable sites was carried out based on distance from the airport, minimum area of the most suitable sites and land use feasibility such as agriculture, forest, built-up areas, road networks, etc. To ensure the longevity of the landfill site, the most suitable sites were filtered by applying a minimum area requirement, providing a lifespan of at least 20 years. Further, the potential sites were filtered by selecting sites between 20 km and 25 km from Ranchi Airport as 20 km was minimum distance as specified in the SWM Rules, 2016. 25 km was selected as maximum distance from airport to ensure that the landfill is not too far from the city, to ensure low transportation costs. A final refinement was done by verifying whether the potential sites were located on barren land. Finally, a sensitivity analysis was conducted to assess the impact on landfill site selection when varying the criteria weights.

2.2.6 Sensitivity analysis

A sensitivity analysis was carried out to understand the changes in model output caused by varying the criteria weights of the input parameters and to identify criteria sensitive to weight changes. The methodology was adopted from Chen et al. (2025)- Spatial sensitivity analysis of multicriteria weights in GIS-based land suitability evaluation to carry out the sensitivity.

The proposed methodology uses the One-At-a-Time (OAT) method, which changes the weight of one input parameter at a time to understand its effect output. A range of weight changes is determined for the study, which is a bounded set of discrete percentage changes referred to as a range of percent change (RPC), from the defined criteria weight. Throughout the RPC, the increment of percent change (IPC), which is the incremental change in each criterion weight during each run within the RPC, is then defined. In this study, the RPC was taken as ± 20 with an IPC of ±10.

For the percent change (pc) of the criteria weight for the main changing criteria, c_m, the calculated weight change, WC (c_m, pc), at a particular pc can be calculated as given in Equation 9

W C c m , p c = W C c m , 0 + W C c m , 0 * p c (9)

Where WC (c_m,0) is the original criteria weight, obtained from FAHP.

But the sum of all criteria weights after every weight change (w_i) should be equal to 1. Hence, the weight of all the other criteria WC (c_i, pc) at a particular pc are also varied proportionally according to WC (c_m, pc) as per the formula given in Equation 10

W C c i , p c = 1 − W C c m , p c * W C c i , 0 / 1 − W C c m , 0 (10)

Where WC (c_i, pc) is the weight of other criteria adjusted proportionally to WC (c_m, pc).

WC (c_i,0) is the original criteria weight of ith criterion obtained from FAHP.

3 Results and discussion

Landfill siting is an important component of rural and urban planning. As per the SWM Rules, 2016, every state has to prepare a master plan for each city in the state or union territory for setting up solid waste disposal and treatment facilities, except for cities with common waste processing facilities. The primary objective of this study was to develop a systematic, holistic, and scientific GIS-FAHP framework to site a long-term landfill for urban sustainability. An AUC analysis under the ROC curve and a sensitivity analysis were carried out to validate the model output and determine the criteria that exerted the greatest sensitivity in the site suitability outcome, respectively.

3.1 Selection of criteria and creation of restriction areas

11 criteria were selected for the study after a survey of literature and considering the statutory requirements based on their hydrological, topographic and infrastructural significance. The significance of these criteria and their respective restriction areas has been discussed in the subsequent section.

3.1.1 Land use land cover

LULC plays a crucial role in deciding the location of a landfill because it shows people’s conventional usage of their natural, geographical, and social resources through time. In this respect, landfill sites can’t be chosen too close to populated and vegetated regions because they may cause public health hazards, lower the value and aesthetics of the area, affect their developmental plans, the fertility of agricultural land, and the biodiversity of the area. Hence, landfills are commonly constructed on barren land rather than fertile agriculturally viable land, water bodies, and forests (Balew et al., 2022; Desta et al., 2025; Khan and Samadder, 2015; Roy et al., 2022). This study has identified five categories of land use: water bodies, built-up areas, vegetation cover or grasslands, agricultural land, and barren land. Distinct weights were allocated to each form of land use. Barren land was considered the most suitable site for landfill and was assigned class 1. Residential areas, agricultural lands, and water bodies are considered unsuitable for landfill construction, and therefore, these criteria were given a higher rating to dismiss these areas for landfilling, as depicted in Table 3.

3.1.2 Slope of the land surface

The slope of a surface location measures the rate of change of elevation and is measured either in degree slope or percent slope (Chang, 2019). The slope of the land is also an important topographic parameter for landfill site selection. Sharp slopes increase the cost of excavation, and hence the appropriateness and desirability for a landfill site decrease with slope steepness. Also, in high-slope areas, surface water runoff is more, the rate of infiltration is less, and this leads to the flow of contaminants over a long distance. In contrast, in a flat terrain, the possibility of groundwater contamination is high (Roy et al., 2022; Singh et al., 2023; Sisay et al., 2025). So, a slope range of 8%–12% has been considered suitable for landfill sites (Khan and Samadder, 2015). In this study, the slope of the area was calculated in degrees using the ‘Slope tool’ in the Surface Analysis Toolbox of ArcGIS.

3.1.3 Distance from settlement

The distance of settlement area/habitation is another important parameter for siting a landfill. Many studies have shown public opposition to creating landfills close to populated areas because they cause health hazards, odour, and groundwater contamination, among others (Islam et al., 2020; Pasalari et al., 2019; Singh et al., 2023). In this respect, the SWM Rules, 2016 have a criterion for setting up landfills at least 200 m away from habitation (MoEFCC, 2016). Accordingly, the minimum distance from the settlement zone has been taken as 200 m; that is, an area within 200 m of the settlement has been given a ranking of 5 (least preferred). Further classes of 200-m intervals were created and ranked. A distance greater than 800 m from habitation has been identified as most preferred and given a ranking of 1.

3.1.4 Distance from river

River is among the most important hydrological criteria for siting landfills. Rivers carry loads, such as rock, sediment, rock, and contaminants, to large distances and deposit them throughout their course. Hence, controlling the amount of toxins and contaminants released into them becomes necessary. At the same time, the discharge of these contaminants adversely affects the aquatic flora and fauna. So, landfill sites have to be created away from rivers to ensure no leachate is discharged into the river, and no plastic and other waste is blown by the wind into river (Ahire et al., 2022; Majid and Mir, 2021; Roy et al., 2022). The SWM rules, 2016, laid down a minimum distance of 100 m from rivers to create a landfill site. In this study, the Euclidean distance tool was used to calculate the distance from the river. This was then reclassified into <100 m, 100–200 m, 200–300m, 300–400m, and >400 m with suitability class as 5, 4, 3,2, and 1, respectively.

3.1.5 Distance from waterbody

Landfill sites cannot be created too close to waterbodies as it adversely impacts aquatic flora and fauna, and human health by leachate discharge, odour, etc. As per the SWM Rules, 2016, landfill sites cannot be situated within a 200 m range from the water bodies (MoEFCC, 2016). The main lakes in the study area are Getalsud Dam, Kanke Dam, Latratu Dam and Hatia Dam. In this study, 200 m around the water body has been given a suitability class 5, and subsequent areas that are greater than 500 m are assigned a suitability class 1.

3.1.6 Distance from groundwater

Landfill generates hazardous leachate and releases gases into the environment. Groundwater pollution may occur from the leachate or due to heavy rainfall. Groundwater contamination from landfill pollutants is a possibility if the site is located in an area where the groundwater level is very close to the surface. Hence, landfills should be constructed at a location where the groundwater level is much below the surface level, where the chances of contamination of subsurface groundwater will be minimized. As per the SWM Rules, 2016, the landfill should be constructed at least 2 m above from highest water table level in that area (MoEFCC, 2016). In this study, a minimum depth of 2 m has been considered. Any location with a groundwater level less than 2 m has been assigned a suitability class 1.

3.1.7 Distance from lineament

Lineament refers to the faults and fractures on the surface of the earth, which may make the rock permeable, allowing water to flow through it (Roy et al., 2022). Construction of landfills is restricted by geological structures such as faults, as leachate may flow through the cracks and contaminate the groundwater. To avoid groundwater contamination, a minimum distance of 100–200 m was considered from faults, joints, fissures, and shear zones to avoid groundwater contamination (Khan and Samadder, 2015). In this study, a 100 m distance from the lineament was given a suitability class of 5. Subsequently, at every 100 m, a suitability class was made. A distance from the lineament of more than 400 m was most preferred with a suitability class 1.

3.1.8 Distance from roads

Roads are an important component of accessibility to landfills. Some studies have provided a higher suitability rank to sites with proximity to roads, emphasising the economic costs of reaching the landfill (Ajibade et al., 2019; Güler and Yomralıoğlu, 2017).Some studies have given a higher suitability ranking to sites farther from road networks owing to environmental impacts, aesthetics, and odour (Ali and Ahmad, 2020; Singh et al., 2023). As per the SWM Rules, 2016, a landfill should be located at least 200 m away from highways (MoEFCC, 2016). In this study, the first restriction zone was created at 200 m, which is any location within 200 m of roadways, and was given a suitability class of 5. Then, zones with 100 m intervals on both sides of roadways were created, i.e. 300 m, 400 m, 500 m, and greater than 500 m from the road, with suitability classes 4, 3, 2, and 1, respectively. A distance beyond 500 m was taken to be most suitable for landfill site construction.

3.1.9 Distance from railway

The railway is an important mode of freight and passenger transport in India. Any hindrance to the railway network has economic costs. So, the distance from the railway track needs to be considered while constructing a landfill, as leachate from the landfill can cause ground subsidence and ruin the aesthetics of the place (Majid and Mir, 2021). In this study, multiple restriction zones were created for railways at 0–100 m, 100–200 m, 200–300 m, 300–400 m, and >400 m with suitability classes 5, 4, 3, 2, and 1 respectively.

3.1.10 Distance from airport

Distance from the airport is another important criterion for siting a landfill. Landfills receive unsegregated organic and inorganic waste, which attracts birds. Proximity to the airport may cause incidents such as a bird hit while taking off or landing an aircraft. This might cause damage to the aircraft and could also lead to a crash. Hence, landfills need to be situated at a suitable distance from the airport (Ghoushchi and Nasiri, 2025; Majid and Mir, 2021) As per SWM rules, 2016 landfills should be at least 20 km away from an airport, unless under special circumstances when it is permitted to be set up between 10 km and 20 km from the airport (MoEFCC, 2016). In this study, the area within 10 km of the airport has been given a suitability class of 5. Construction of a landfill is the least preferred within this boundary. Suitability class 4 has been assigned to the area between 10 km and 20 km from the airport. The area beyond 20 km becomes cost-effective for siting a landfill. So, the area between 20 and 30 km, 30–40 km, and beyond 40 km has been assigned suitability classes 1, 2, and 3, respectively.

3.1.11 Soil

Soil permeability and porosity are important deciding variables to reduce the chances of groundwater pollution. Permeable and porous soil allows leachate to percolate to the groundwater at a higher rate. Clayey Soils have low permeability, and bedrock with a high proportion of soil with poor permeability are perfect for landfill site selection because high clay content in soil has a low permeability, which decreases leachate entering into the earth compared to other soil types (Ali and Ahmad, 2020; Sisay et al., 2025). The study area has 4 major soil types-clayey loam, sandy clay, sandy clay loam soils, and loam, that have been rated as 1, 2 3, and 4, respectively for the study.

3.2 Criteria weights and landfill suitability map

The pairwise comparison matrix created for each criterion assigned its relative importance with respect to other criteria for final landfill siting, as depicted in Table 2. The consistency ratio of this pairwise comparison matrix was 0.064, which is within the threshold of 0.1. Weights were obtained for each layer by applying Buckley’s geometric mean method. Table 3 shows that LULC has been awarded the highest weight among all criteria in the study. This means that LULC has the most effect on determining the landfill site, which can be attributed to the higher importance placed on LULC relative to other parameters in the comparison matrix. This result is also aligned with the literature, because LULC has been reported in many studies as the most important criterion (Bera et al., 2018; Khan and Samadder, 2015). LULC is followed by slope, distance from settlement, distance from rivers, and distance from surface water bodies in terms of the criteria weight.

The Weighted Overlay tool in ArcGIS was used to generate the landfill suitability map, with 4 suitability classes classified using natural breaks (Jenks). These classes were named as Class 1, Class 2, Class 3 and Class 4. Figure 4 shows the landfill suitability map where Class 1 represented the most suitable sites for landfill siting, and Class 4 represented unsuitable for landfill. Classes 2 and 3 were assigned as moderately suitable and less suitable respectively. The results show that the area with suitability class 1 is around 151 km², covering 3% of the total area and encompassing 614 sites. Suitability classes 2, 3, and 4 have an area of about 3,566 km² (70% of the total area), 1,298 km² (25.5% of the total area), and 81 km² (1.6% of the total area), respectively. This implies that most areas in the district are moderately suitable, but sufficient sites, i.e., 614 sites in suitability class 1, can be used to develop as landfills.

These results also align with studies conducted locally and globally regarding the percentage of the landfill area classified as most suitable in the landfill suitability map. These results also align with studies carried out locally and globally in terms of the percentage of area classified as most suitable in the landfill suitability map. Balew et al. (2022) employed the GIS with AHP to site landfill sites in Robe town, Ethiopia and found that 6.08% of the total study area was very highly suitable for landfill siting in a 5-class suitability map. Similarly, Tripathi et al. (2022) in their study combined GIS with AHP and FAHP and found that 18.78% and 14.33% of the total study area was suitable for landfill siting in Prayagraj, India. Roy et al. (2022) assessed landfill site suitability using GIS- AHP framework in Siliguri, West Bengal and found 3.48% of the total area as highly suitable for landfill siting. Sisay et al. (2025) also carried out an GIS- AHP study to identify landfill sites in Gimba town, Ethiopia and reported that 1.6% and 34.6% of the total study area was very highly suitable and highly suitable respectively, for landfill site selection.

3.3 Potential landfill sites for waste management

This study reevaluated all the identified sites for its suitability for solid waste management, focusing on sites with optimal size, infrastructural consideration and socioeconomic factors. First, a minimum landfill area was identified to ensure a landfill operation time of 20 years. A larger landfill site lowers the necessity for repeated site selection and costs associated with design and operation (Sisay et al., 2025). Section 2.2.4 of the methodology describes the steps followed to derive this minimum area. Accordingly, 27 ha was calculated as the minimum area required for landfill site selection. As a result, the number of prospective landfill sites was lowered from 614 to 191, located in the Ranchi district. This is in a line with literature that have re-evaluated their most suitable landfill sites derived from landfill suitability map to refine the list of potential sites (Singh et al., 2023; Sisay et al., 2025).

Another important criterion is the distance of landfill site from the city as transportation cost is a function of distance. Transportation coat would be very high for a long distant landfill whereas the landfill may become a nuisance to city’s infrastructure if located too close. In this regard, the first buffer was created at 20 km as per SWM Rules 2016 mandate that no landfill site be located within 20 km from an airport. Hence, the distance from the city was calculated and the airport at Ranchi was taken as the epicentre. The second buffer was created at 25 km so that the transportation cost, operational cost and capital cost do not increase rapidly. So, the number of potential sites reduced from 191 to 35 as depicted in Figure 5a. These 35 sites are prospective sites for landfill that are close to Ranchi city. The area of the sites ranged from 27.6 ha to 941 ha, implying that vast areas of continuous land are available, which can be converted to engineered landfills.

The final re-evaluation was carried out by verifying the 35 sites for their land use, as sustainable land management deems land under natural vegetation and agricultural usage unfit for landfill siting. Each site was verified for its land use and land cover using Google Earth Pro, and 4 sites lying in Barren land were selected, as shown in Figure 5b. Table 4 provides the details of the 4 identified sites that could be developed as landfills.

TABLE 4. Identified landfill sites.

Validation of model output is a crucial step to determine the model’s ability to yield precise and comprehensive predictions. The Receiver Operating Characteristic (ROC) is an important tool which helps to identify the prediction accuracy of the models. To validate the model outputs of this study, 70 points were identified in the study area through field visits and Google Earth Application and AUC- ROC curve was employed to validate the model. The ROC curve yielded an AUC of 0.91 which reflect strong predictive and discriminatory capability of model, confirming the robustness of GIS- FAHP framework used for this study.

3.4 Sensitivity analysis

In this study, an RPC of ±20% and an IPC of ±10% were applied to all 11 criteria weights to study the changes of the cell in the evaluation map, as described in Section 2.2.6. of the Methodology. Within the range of −20 to +20 with an IPC of 10, 4 evaluation runs were obtained for each criterion. Table 5 lists the changes in weight of the 3 criteria that have the maximum weight as per the FAHP calculation, which includes LULC, slope, and distance from the settlement zone, with 21, 17, and 13 as their respective weights. Figure 6 represents the change in cells between the suitability classes when the criteria weights of the 11 criteria were changed by ±10% and ±20%

TABLE 5. Change in criteria weights of the 3 most important criteria.

After the analysis of the results obtained from 44 simulation runs, it was observed that Distance from the settlement has the highest sensitivity, whereas distance from railway and soil type have the lowest sensitivity. Also, when the criteria weight for distance from the settlement was increased by 20%, the maximum change in cells in the evaluation map (3,022 cells) was observed. This change in criteria weight reduced the area in suitability class 1 by 0.5% (from 15,098.49 ha to 15,020.8 ha). Also, 2,902 cells of the total 3,022 cells changed from moderately suitable to less suitable. The spatial distribution of this change was observed in cells closer to settlement. Other criteria that showed high sensitivity were distance from waterbody and distance from river when their criteria weights were reduced by 20%.

In this respect, it has been seen that the criteria with the most weight may not be the most sensitive. This is because the criteria weight refers to the relative weight of each criterion to other criteria, whereas sensitivity deals with the change in decision outcome when the weight of a criterion is varied. Hence, a criterion with less weight may be more sensitive owing to its characteristic and the dependence of the model output on the criterion.

Distance from settlement has the third-highest criteria weight in the order of criteria weights, yet it was observed to be the most sensitive of all the criteria weights. This can be attributed to the distribution of population throughout the district. As the population is widely distributed, any change in the criteria weight has larger changes in the suitability class of the cells, and subsequently, the criterion is more sensitive.

Further, to understand the spatial implications of the sensitivity analysis, potential landfill sites were identified using modified the criteria weights obtained from the analysis. Potential landfill sites thus identified were subjected to the conditions: a) have a suitability class 1, b) lie between 20 km and 25 km from the airport, and c) have a minimum area of 27 ha. Table 6 lists the total number of landfills identified in the entire district and the narrowed-down landfills located within a 20 km and 25 km buffer.

TABLE 6. Change in landfill location as per sensitivity analysis.

After the spatial location of landfills, it can be observed that the largest change in the number of identified landfills with suitability class 1 and minimum area of 27 ha in Ranchi district, was for Slope, with a 10% increase in criteria weight, wherein the number of identified landfills was 83 and a change of 108 from the original landfill sites. This was followed by LULC, which had a 20% decrease in criteria weight; slope, with a 20% increase in criteria weight; and distance from settlement, with a 20% decrease in criteria weight, all of which identified 84 landfill sites in the entire district of Ranchi. In terms of the number of landfill sites identified between 20 km and 25 km, the changes were similar to the changes for the entire district.

These maps have shown that after applying the 3 conditions mentioned earlier, slope and distance from settlement show the most sensitivity in identifying landfill sites. Increasing the criteria weight of slope by 10% and 20% and decreasing the criteria weight of distance from settlement by 20% has shown a drastic decrease in the number of identified landfills from 35 to 8 and 8, respectively.

4 Conclusion

GIS-FAHP framework could be a useful tool to identify the landfill sites for sustainable, economic and scientific land management. From this study, it is found that LULC is the most important criterion with a weight of 21%, followed by slope and distance from settlement with 17% and 13% weight respectively for landfill site in Ranchi. After overlaying all the criteria maps, it is observed that around 3% of the total area of the district is most suitable for siting a landfill.

For a landfill operation time of 20 years, the minimum area of the landfill is estimated as 27 ha, which reduced the number of potential sites to 191. After re-evaluating the sites by applying 20–25 km distance criteria, a total of 35 sites is identified, with areas ranging from 27.6 ha to 941 ha. Finally, 4 sites are identified after verifying their land use as barren land.

Sensitivity analysis indicated that distance from the settlement is the most sensitive criterion, showing a change in cell category of 3,022 when the criteria weight is increased by 20%. When the criteria weights of distance from waterbody and distance from river are reduced by 20%, there is change in cell category of 2,896 and 2,890 respectively.

This work may offer insights for policymakers and solid waste management departments in identifying suitable locations for engineered landfills. It may also serve as a useful reference for students, academics, and other readers in assessing land suitability for applications such as landfill siting, waste-to-energy facilities, and recycling plants.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Suneela Sardar, University of Lincoln, United Kingdom

Reviewed by: Sujoy Kumar Malo, University of North Bengal, India

MD SK, Rajendra University, India