Decoding e-waste challenges with hybrid AHP and ISM Model approach: An initiative towards a cleaner future in India

2.1 e-waste challenges identified from literature

13 key challenges were identified that are a hindrance to e-waste alleviation through management in India. A brief overview of the various challenges that pose a threat to e-waste mitigation and their current status is as follows:

Εε1: Lack of Infrastructure [LI]

The e-waste management practices used in developed countries cannot be implemented in India due to unavailability of adequate number of industries to address the initiation of the mitigation process through 6R and circular economy implementation. Thus, the governance, investors and all stakeholders need to work on a vision to capacitate India to unlock its infrastructural prowess. Presently, it is a hindrance to be dealt with firmly (Bagwan, 2024).

Εε2: Lack of Knowledge and Skills [LKS]

The recent decade brought the concept of e-waste mitigation to India. As obvious, it got cladded with a fundamental issue of exposure to the novel problem and approach to solve them. No knowledge and skills or means to acquire them is still a critical root-level hindrance to initiate the efforts towards the success of this hazard alleviation (Sundar et al., 2023).

Εε3: Financial Constraints [FC]

The cost aspects of introducing any new idea such as e-waste mitigation has its own problems and constraints amidst the potential catastrophe it possess globally if not addressed with diligence. This constraint of finance is due to lack of motivation of investors, government support initiatives and diversion of funds to fuel this mission on an urgent basis. Hence, it becomes a basic challenge to be eradicated for the success of this mitigation initiative (Awino and Apitz, 2024).

Εε4: Information Constraints [IC]

Lack of public awareness and data accessibility issues due to lack of transparency or security concerns escalate the problems of resolving the e-waste barriers and is a result of the cumulative effect of basic barriers identified. It aggravates further into low customer awareness which triggers the collapse of the entire endeavor. Hence, it is a vital challenge to be addressed to initiate the mitigation and transformation process towards sustainability (Mohideen et al., 2024).

Εε5: Fluctuating Supply [FS]

Constraints like lack of information and customers’ attitude leads to an unstable and fluctuating supply input to the production process of e-waste, which eventually increases costs and hence in combination aggravates the challenge of e-waste mitigation in India (Debnath et al., 2023). It is an enormous barrier for sustenance of infrastructural facilities to initialize the mitigation mission (He et al., 2024).

Εε6: Immature R&D Practices [IRDP]

India is only over a decade young into e-waste mitigation management issues. The know-how to initiate any new thought system has not availed adequate timeframe leading to immature R&D practices with execution flaws leading to a hindrance in achieving the e-waste sustainability goal (Neves et al., 2024).

Εε7: Logistics & Transportation practices [LTC]

Logistics and transportation being the backbone of any supply chain, in mitigating e-waste it has a vital role to play. In this context, the lack of industries incorporating 6R and inadequate demand and supply act as precursors to increase costs of operation, hence, enhancing the hurdles towards mitigating e-wastes in India (Mallick et al., 2023).

Εε8: Incoherent Legal and Policy constraints [ILP]

In India, laws do exist for e-waste alleviation but is redundant as availability of suitable industries or big companies imbibing circular economy is sparse. Further, appropriate policies can’t be defined without a threshold ambience of certain constraints’ alleviation as precursors. This trade-off is a serious concern to be addressed (Faibil et al., 2023).

Εε9: Illegal practices [IP]

As the dominating informal and unorganized sector operates in India for e-waste recycling in a very gross way, to maximize profits they often retort to illegal practices such as import and dumping, and unethical approaches to recycle that enhance toxicity of the environment thereby hindering the core goal of achieving e-waste sustainability (Rauf, 2024).

Εε10: Lack of customer’s attitude [LCA]

The most vital precursor challenge in e-waste transformation into sustainable paradigms in the customer-driven market is their attitude towards 6R products which include taboos, social status quo issues, psychological resistance due to older ideals etc. It is one of the most powerful barriers to be overcome for initiation mechanism of e-waste mitigation (Gaur et al., 2024).

Εε11: Lack of Sustainable Design Practices [LSDP]

Lack of sustainable practices act as a hindrance as an overall sum of all the other 12 challenges identified in the literature. Design is a combination of numerous sustainable aspects and a right balance blending is very difficult to achieve. Under these initial years of Indian struggles to alleviate e-waste challenges, it is a goal of colossal magnitude to be achieved (Nandy et al., 2022).

Εε12: Lack of Safety constraints [LSaC]

Constraints like lack of knowledge and skills, finance, and awareness of workforce cumulatively add-up to safety issues through various aspects such as non-mastery in work, unaffordability to provide mechanisms and measures, and human errors respectively. The first two of these aspects are constraint-driven and can be addressed towards minimization which is yet to be achieved. This hinders and demotivates the workforce as risk increases, hence, a vital challenge to e-waste mitigation prospects (Bimpong et al., 2024).

Εε13: Improper Coordination & accessibility [ICA]

Coordination of unstructured workforce and their accessibility to various resources such as skills and knowledge levels, and information brings forth a hazy picture of their work prospects psychologically leading to demotivation and brooding conflicts. It leads to failure of the goal of sustainable e-waste management achievements and hence an important challenge to be conquered (Leclerc and Badami, 2024).

3.1 Data collection

The authors, through a rigorous literature survey identified 13 hindrances to e-waste management initiation in Indian context. These challenges were to be validated by the experts to identify the significant ones for management approach and discarding the others. Two robust MCDM techniques viz., Analytic Hierarchy Process (AHP) and Interpretive Structural Modeling (ISM) were employed, the former to deduce weightage of challenges along with coherency of experts’ opinions as a validity check while the latter to identify levels of challenges identified which are to be resolved from level 7 (Short-term focus/basic or root barriers) and gradually building to the most complex (Long-term focus/advanced barrier) at level 1 through an ISM model. The need for data collection arose as both these MCDM techniques get initiated through meticulous opinions of experts which they cultivated through their work experiences and observations (Podvezko, 2009).

For AHP as well as ISM, a total of same 15 experts responded at one point of time, 9 from research/academia and 6 from industries with e-waste management experience. For evaluating preferences in AHP, a 9-point scale was used which reads as: 1: Equal importance; 2: Equal to moderate importance; 3: Moderate importance; 4: Moderate to strong importance; 5: Strong importance; 6: Strong to very strong importance; 7: Very strong importance; 8: Very to extreme strong importance; 9: Extreme importance. 16 challenges identified through literature survey when validated through AHP led to a final 13 barriers qualified for managerial analysis. These are coded as h-1 to h-13 for analysis purposes. The 3 insignificant challenges discarded involve unstable composition (of e-waste products), uneven escalation of e-waste generation, and weak stakeholder communication/collaboration. For ISM model the experts gave their valuable insights through a 5-point Likert scale interpreted as: 1: No importance; 2: Low importance; 3: Importance; 4: High importance; 5: Very High Importance. The number of experts for both AHP and ISM are within appropriate limits. AHP gives better results with greater sample size while for ISM, the number of experts can start from as low as 3. These 13 challenges finalized whose alleviation would mitigate e-waste were to be put into a structure or sequence (levels) through a model, which would help resolve them from basic to advanced challenges.

3.2 AHP

Saaty (1990) developed a practical method for decision makers to gain insights to numerous subjective, complicated, and conflicting criteria through a robust MCDM technique, the Analytic Hierarchy Process (AHP). The steps of AHP are elucidated (Khoshand et al., 2019) in the following sub-sections:

3.2.1 Pairwise matrix

The insights of experts collected through the 9-point scale which occupy the upper-triangular portion of the (n x n) matrix showcases a hindrances’ relationship with other (n-1) hindrances. The diagonal elements are valued at 1 as they refer to the same hindrance. The lower triangular portion are given reciprocal values of experts’ opinions i.e. if a₁₂ = 3, then a₂₁ = 1/3 as it is a pairwise influence comparision. 15 experts provided opinions as a matrix whose cell-wise matrix were averaged to find average pairwise matrix. It forms the base for the entire calculations involved in AHP. The column sums of each of the 13 columns were summated.

3.2.2 Normalized weightage matrix

Each column sum obtained in pairwise matrix is divided in their respective column in pairwise matrix to obtain the normalized matrix. Then, average of each row of normalized matrix results in weightage of each criterion which is ranked accordingly.

3.2.3 Model robustness

The criteria weights are multiplied to each cell of column and row wise in two steps. The first step involves multiplying criteria weights of each hindrance to the entire column of the same hindrance, similarly, for all hindrances in the pairwise matrix ( Table 5). Then, each criteria weight for the hindrance is multiplied row-wise to obtain the final cell values. The weighted sum of each hindrance is obtained by row summation of each hindrance of Table 6. X represents a parameter of ratio between weighted sums for each hindrance to respective criteria weight.

This value of X is used to calculate Consistency Index (CI) and Consistency Ratio (CR) through the following formulas:

The general thumb-rule for data consistency in AHP is “closer the value of C.R. to zero, greater the data consistency, and, the acceptable limits of consistency for C.R. value is less than 0.10”.

3.3 ISM

Interpretive Structural Modeling (ISM) is a popular and proven MCDM technique with data-driven initiation; the data being experts’ opinions in a particular field of expertise. The views from experts are collected for validation after identifying challenges through vast literature survey. The VAXO scale for influence (Singhal et al., 2018) is employed to examine a pairwise relation through a self-interaction matrix. The sub-section 3.3.1 to 3.3.5 invokes the steps for ISM model building and analysis purposes as follows:

3.3.1 SSIM

15 industry experts associated with waste management provided hawk-eye views which enriched the authors to extrapolate the contextual relationship among the e-waste challenges. The nature of the relationship between two factors (x and y) is established using VAXO conditions where:

V: if x influences y; A: if y influences x; X: if x and y influence each other; O: if x and y have no relation.

The SSIM is a precursor to obtaining the reachability matrix.

3.3.2 Reachability matrix

It is an outcome of SSIM being converted into a 0 and 1 binary value matrix where if a factor (x) influences factor (y), then the value is 1, otherwise 0. Under this condition, for a factor set [x to y] or [y to x] influence depicted as [x, y] or [y, x] matrix cells position for each of VAXO are:

Thus, the binary reachability matrix is obtained in Table 8 in which adding rows gives a “Driving Power” column while adding columns of the matrix gives a row “Dependence Power”. These are plotted in as y and x axes in MICMAC analysis (section 3.3.5).

3.3.3 Level partition

Table 9 represents the level partition of hindrances where values of each cell of reachability matrix (with values 1) in rows and columns are considered for reachability set (R-set) and antecedent set (A-set) respectively. Rows with values 1, represent that the hindrance h_x (x = 1, 2 …, 13; each row having one x value) i.e. Εε1 for instance, influences how many other hindrances whereas column values with 1 implies how many hindrances get influenced by Εε1.

Further, an Intersection set (I-set) is deduced which showcases the common challenges in R-set and A-set.

The procedure for level indexing after the R, A, and I-sets elements identification is encapsulated as:

3.3.4 ISM model

The ISM model which uses a diagraph, showing hindrances as nodes or vertices and joined by edges, exhibits interdependence between them. The basic assumption from an ISM model is transitivity i.e. if Εε1 influences Εε2, and Εε2 influences Εε3 implies Εε1 will influence Εε3 transitively. The authors have used a Floyd-Marshall algorithm using Python 3.0 code with initial reachability matrix as input to remove transitivity and give final reachability matrix without transitivity as output which is used to generate the model. The model illuminates the structural relationship among the factors. Thus, ISM converts a complex set of numerous factors to a well-established interrelationship for clarity of the research context as a model (Chakraborty et al., 2023). It represents a clear view of the issue by focusing on the head-on and ancillary relationship between the challenges. It combines the experts’ opinions and their experiential depths in a superior organized way.

3.3.5 MICMAC analysis

The driving power and dependence power of each challenge from reachability matrix are plotted in a quadrant chart where the origin is average of 0 and highest dependence power i.e. 13 in x-axis and in y-axis it is average of 0 and 10 dependence power implying origin to be at (6.5,5). His plot is referred to as MICMAC (Matriced’ Impacts croises-multiplication appliqu’e an un Classement) analysis.

Figure 1 shows the framework of e-waste management implementation hindrance research design procedures.

4.1 AHP calculations

The AHP matrix calculations to final weightage and ranks of challenges to e-waste management implementation are represented from Table 3 to Table 6.

4.2 ISM calculations

The Tables from 7 to 9 showcase ISM matrix calculations leading to level partition table for model building.

5.1 AHP: Hindrance weightage and data validation observations

The second research objective (RQ2) to prioritize or rank the e-waste challenges through their weightage calculations are accomplished by AHP.

The weightage of hindrances to e-waste implementation are ranked as:

It shows the importance of various challenges to e-waste implementation.

Lack of knowledge and skills tops the weightage with 12% followed by lack of infrastructure (10%). Financial and information constraints are equally important with 9% weightage each, placed in third position. The fourth position is a tie between immature R&D practices, and Logistics and Transportation practices (8%). Fluctuating supply (7%) holds the fifth position followed by equally important challenges viz., Illegal practices and lack of customers’ attitude (4% each). For the seventh position i.e. the least important among challenges it has three challenges at 3% each namely, lack of safety constraints, improper coordination and accessibility, and lack of design practices.

The results obtained for expert opinion data validation and consistency is as per the rules (Section 3.2.3; Equations (1), (2), (3)).

As noticed, CR is less than 0.10 and its value is 0.053 tending towards zero, thereby showcasing highly consistent opinions of experts.

5.2 ISM: Model and MICMAC insights

The ISM model elucidates very precisely the origin/present state of hindrances to e-waste mitigation and how in 7 levels reach to the cumulative hurdle of LSDP. Alternately, it can be viewed that the model lucidly explains the barriers which kindles aspirations for a revival strategy to overcome them. The ISM model delivers a structure of hindrances to e-waste implementation. The summary of the model characteristics ( Figure 2) are discussed below.

Figure 2. ISM model.

The ISM model insights can be envisioned in 7 levels with the seventh one signifying the short-term focus or the immediate problems/challenges to be addressed. These are lack of infrastructure (LI), lack of knowledge and skills (LKS), financial constraints (FC), and lack of logistics and transportation practices (LTC). Level 6 represents lack of customers’ attitude (LCA) which is triggered by hindrances of L-7. This aggravates the lack of information constraints (IC) at L-5 leading to illegal practices (IP), incoherent legal and policy constraints (ILP), and improper coordination and accessibility (ICA). These three L-4 challenges are interconnected and complement each other to build an “envelope of hindrances” encompassing the mitigation of e-waste solutions. It further becomes formidable with L-3 barriers of fluctuating supply (FS) and immature R&D practices (IRDP) which complement each other. These, in union, lead to lack of safety constraints (L-2) which finally boils down to the L-1 hindrance of lack of sustainable design practices (LSDP).

The key insights from the model can be encapsulated as follows:

Thus, L-7, L-6, and L-5 hindrances are the independent or driving factors to be given high priority for resolution initiation.

Hindrances Εε8_, Εε9, Εε13 viz., IP, ILP, and LCA have interconnectivity creating a “sphere of hindrances” making their alleviation difficult, thus, called unstable or linkage factors.

Figure 3 represents MICMAC analysis, which classifies the thirteen identified challenges into four sections, such as drivers, dependents, linkage and autonomous. Table 10 shows the different categories of challenges.

Figure 3. MICMAC analysis of e-waste challenges.

There is a unique observation between the results of AHP and ISM owing to the fact that the same 15 experts with their acumen provided their opinions at the same point of time. There is a horizontal trend observed that long-term focus or dependence factors are the least important hindrances, 8 out of 13 challenges of e-waste management implementation in Indian context corresponds to a trend that “the driving factors (short-term focus) to dependence factors (long-term focus) in ISM correspond to most weightage to least weightage (importance) in AHP”. Also, a narrow range between all hindrances signifies that the ISM model is fragile and susceptible to readjustments for some hindrances. Therefore, it would require proactive approach to alleviate these e-waste management implementation barriers, sooner the better.