Social Life Cycle Assessment Methodology to Capture “More-Good” and “Less-Bad” Social Impacts—Part 1: A Methodological Framework

Abstract

Social life cycle assessment (SLCA) systematically assesses the social impacts of the entire life cycle of a product system or service that stretches from extraction and processing of raw material to recycling and final disposal. Most SLCA techniques highlight negative impacts and their reductions, while positive social impacts and their increments have received less attention. Positive social impacts highlight chances for improving human well-being and present a complete picture of a product’s overall social impact. The literature shows that norms for defining positive impacts and methodologies for assessing them are not yet fully established and retain lacunae, which can lead to conflicts in the usage of the term “positive impacts”. Therefore, we develop a novel SLCA methodology that can straightforwardly distinguish between the “good” and “bad” social state at the subcategories in the latest version of methodological sheets for SLCA. Here, we refrain from using the terms “positive” and “negative” as those terms retain scattered consensus; instead, we use the fresh terms “good” and “bad”, which are simpler to understand. To describe the positive changes in good and bad states, we introduce two new terms into SLCA: “more-good” (improvements within the good domain) and “less-bad” (improvements within the bad domain). Good and bad social domains are distinguished using compliance levels (e.g., industry standards), referred to as baseline requirements. Social impacts were evaluated using the social performance index (SPI). The SPI is computed by multiplying social performance levels with working hours at the factory/company level. Social performance levels are evaluated using a decision tree and a systematically proposed set of indicators representing basic requirements and good and bad domains of each subcategory. Working hours were used as an activity variable, estimated using a working hour model. This enables the application of the SPI across the supply chain of a product by linking social impacts to the time spent on each activity.

1. Introduction

Social life cycle assessment (SLCA) is a popular tool to systematically assess the social impacts of the entire life cycle of a product system or service that stretches from extraction and processing of raw material to recycling and final disposal. Although the roots of SLCA date to 1996 [1], a comprehensive approach was available in 2009, termed ‘UNEP/SETAC Guidelines for Social Life Cycle Assessment of Products’ [2]. Building upon that, various researchers developed the methodology further up to the latest version of ‘Guidelines for social life cycle assessment of products and organizations 2020’ [3]. Finally, a standardized framework for SLCA, ISO 14075, was published in November 2024 [4].

Social impacts in SLCA are assessed concerning an area of protection (AOP), i.e., human well-being [5]. Human well-being describes the state of an individual’s life situation [3]. The SLCA guidelines link human well-being to six stakeholders: workers, the local community, society, children, consumers, and value chain actors [6]. These stakeholder categories are further engaged with 40 subcategories describing different impact areas where activities in the value chain may impact on the human well-being of the related stakeholder [5]. The impacts of these subcategories are assessed using inventory indicators supplemented by the guidelines.

Following the path of environmental life cycle assessment (ELCA), most SLCAs to date have reported negative social impacts (damages) and their reductions (e.g., Manik et al. [7], Hosseinijou et al. [8], Lenzo et al. [9], Yıldız-Geyhan et al. [10], and Dunuwila et al. [11]) or conducted generic hotspot assessments reporting potential negative social impacts (e.g., Teah et al. [12], Toboso-Chavero et al. [13], Holger et al. [14], Dunuwila et al. [15], and Gong et al. [16]).

However, reporting positive impacts is barely discussed and deemed necessary [17,18,19]. Positive social impacts disclose avenues to improve human well-being [5]; hence, they should receive more attention to provide a complete picture of the overall social impact of a product or service and to facilitate further improvements in human well-being.

The definition of positive social impacts has been a topic of debate among researchers, who follow different norms to describe them. For instance, Norris [20] regarded improved health benefits due to economic development in product life cycles as positive impacts. Further, Traverso et al. [21] stated that improved quality of life through wages, holidays, and employee contracts could lead to positive impacts. Nevertheless, these remunerations may be considered essential mandates established by the law to protect employees [5]. As per Baumann et al. [22] and Ekener-Petersen and Moberg [23], ”the utility of goods” can create positive impacts. While “utility” may enhance “well-being” according to economic terminology by incorporating win-win scenarios where all parties involved benefit [24], the notion of positive impacts extends beyond that. Because focusing solely on “utility” and “well-being” as defined by traditional economics may not capture the full scope of positive changes that can occur. Social handprint is another approach to capturing positive social impacts. Being the opposite of “social footprint”, it accounts for the sum of positive changes that someone causes outside the scope of the footprint. When handprint overrides the footprint, net positive social impacts occur [25]. However, the social handprint is assessed by referring to the business as usual (BAU) scenario that tends to vary temporally and geographically; hence, it does not present a universal platform for comparison-based decision-making. Meanwhile, Ekener et al. [5] categorized the subcategories in the previous UNEP/SETAC Code of Practice for SLCA into positive and negative groups based on the norm that positive impacts relate to additional benefits such as local employment creation, improved infrastructure, and so on. However, further improvements were required before putting this approach into practice. Meanwhile, Franze and Ciroth [26] regarded the low level of negative social performance as resulting in positive social impacts. They included this norm in a five-point reference scale to measure a set of social indicators, where a score of one was given for highly positive social performance, while a score of five was given for highly negative social performance. The absence of negative impacts is not necessarily “positive”; instead, it is neutral [5]. Because if production is absent, issues such as forced labor and child labor may become irrelevant while all the benefits of positive aspects are nullified. A similar approach was present in Ekener-Petersen and Finnveden [27]; for instance, the subcategory “social benefit/security” was assessed by the degree of social security expenditure (spending as a % of GDP), where higher expenditure was regarded as resulting in positive impacts. Here, a question arises where a sufficient or neutral level of expenditure exists. Ramirez et al. [28] and Hannouf and Assefa [29] attempted to address this issue by defining a sufficiency level (i.e., termed basic requirement) for each subcategory in the previous version of UNEP/SETAC SLCA guidelines. Mainly, labor laws, international standards, and UNEP/SETAC SLCA guidelines were referred to in this regard. Product Social Impact Assessment refers to “basic requirement” as a “generally acceptable situation” and includes it in a reference scale as a boundary separating the positive and negative performances [30]. Attempts of a similar nature are evident in Corona et al. [31], Osorio-Tejada et al. [32], and Garfi [33]. The latest version of UNEP/SETAC SLCA guidelines delivers a more detailed overview of positive impacts wherein it defines three types of positive impacts, i.e., Type-A, -B, and -C. Type-A positive impacts were described as the social performance going beyond business as usual, which is often at the compliance level but sometimes below or above compliance. Type A somewhat resembles the definition given in the old version of UNEP/SETAC SLCA guidelines—“performance surpassing compliance stipulated by laws, international standards, etc.” [2]. Type B represents the positive social impacts generated through the additional beneficiaries, such as employment, capacity building, or improved infrastructure of a product or company. Type C describes the positive social impacts prompted by intrinsic characteristics of the product utility, for instance, vaccines or effluent treatment plants aimed at improving the well-being of the people [3]. While ISO 14075 provides a standard framework for social life cycle assessment, it offers limited specific guidance on assessing positive social impacts [4].

The literature reveals a lack of established norms and methodologies for defining and assessing “positive impacts” in SLCA, which can lead to inconsistencies in the application of the term. To address this gap, we develop a novel SLCA methodology that aims to clearly distinguish between “good” and “bad” social states. In this framework, we move away from the terms “positive” and “negative”, which, despite their widespread use, lack a consistent consensus in the SLCA literature. Instead, we introduce the terms “good” and “bad” to denote desirable and undesirable social outcomes, respectively, offering a more readily understandable and intuitively accessible distinction. Furthermore, we introduce the terms “more-good” and “less-bad” (defined in Section 2.1) to describe positive changes within the “good” and “bad” states in each subcategory following an intervention or improvement. Our methodology is designed to be comprehensive, encompassing all subcategories outlined in the latest version of the methodological sheets for SLCA, and may remain adaptable across diverse geographical and political contexts.

2. Materials and Methods

2.1. More-Good” and “Less-Bad” Concepts

We introduce novel terms “more-good” and “less-bad” to SLCA to describe the positive changes in good and bad social states (see Figure 1). We try to separate the good and bad social state of an organization involved in a production with a neutral level, termed basic requirement at the subcategory level (see Section 2.2 for more details). Specifically, the basic requirements can be defined as the criteria referring to either generally accepted international norms where no bad social impacts occur or the situation where organizational activities do not affect stakeholders [28,29,30]. Performing better than the basic requirement leads to a good social state, which ultimately yields good social impacts. Unfulfillment of basic requirement(s) results in a bad social state leading to bad social impacts. That is, the domains above and below begin to represent the organization’s good and bad social states, respectively. Suppose the organization’s current condition is in the good domain and improves further toward the good direction, or the current condition is in the bad domain and improves into the good domain, crossing the basic requirement, following a decision. Here we call these movements “more-good”. If the current social condition of the organization is within the bad domain and improves toward the basic requirement after the introduction of a decision, this movement is deemed “less-bad”. As a whole, preceding movements can be coined positive changes. Additionally, “less good” and “more-bad” are available. The former can occur when the current condition remains in good domain and plunges toward the basic requirement. The latter is evident as the current condition remains in the bad domain and worsens, or the current condition is in the good domain and moves into the bad domain, crossing the basic requirement with the insertion of the decision. Overall, these movements can be deemed negative changes.

2.2. Concept of the Basic Requirement

Generally, to date, these basic requirements are defined per organizational standards (ILO guidelines), international or local agreements, UNEP/SETAC SLCA guidelines, and other literature [28,29]. In studies like Ramirez et al. [28] and Hannouf and Assefa [29], the basic requirement refers to the presence of a policy or a single practice that complies with the industrial standards or agreements. However, we argue that merely “the presence of a policy” cannot be at a neutral level, as a factory with an acceptable policy may not implement the policy guidelines in real situations.

Further, we argue that sometimes “multiple” basic requirements may exist for some subcategories to reach the neutral level due to their multidimensional nature. A single basic requirement may be insufficient for complex issues with multiple aspects, potentially oversimplifying the reality and failing to capture the detailed characteristics of social performance. By employing basic requirements, we can achieve a more holistic and accurate representation of the subcategory’s social impact, allowing for a more granular and comprehensive assessment of its various dimensions. Saling et al. [30] defined multiple basic requirements for some subcategories in their framework. However, they do not explicitly discuss why they use multiple basic requirements. On the contrary, for instance, in our case, the basic requirements of freedom of association and collective bargaining are, “Do the workers have a proper mechanism to promote, defend and negotiate their respective interests, collectively and freely? (At least one of the activities listed below is present and/or workers are free to form them) (e.g., Unions and other collective activities), Are workers free to join the above activities of their choosing? Are there no incidents reported to prevent the above activities?” Here, we believe that if one basic requirement is not fulfilled, the organization cannot reach the neutral level. These basic requirements are defined based on ILO guidelines (No. 87) [34], UNEP/SETAC methodological sheets for subcategories in SLCA 2021 [6], and product social impact assessment framework [30]. The list of basic requirements per each subcategory and their supporting literature is mentioned in the Supplementary Material.

2.3. Methodological Framework

Our methodology evaluates subcategories; hence is a subcategory assessment method (SAM), of which the impact assessment phase belongs to the Type-1 impact assessment group (see Section 2.3.3 for more details on Type-1 methods).

2.3.1. Goal and Scope Definition

This sub-step was inserted to define the study’s goal, scope, and functional unit. The use of the functional unit in SLCA has been controversial, as the intangible social impacts remain challenging to link to the physical flows. Therefore, the functional unit has been defined nominally in many SLCA methodologies, including SAM [28]. We try to address this issue by relating the functional unit to the activity variable, “working hours”, which is more inherent to labor conditions [35] (for more details on the activity variable, see Section 2.3.3). The time scope adhered to for the evaluation of inventory indicators is set as three years [30].

2.3.2. Inventory Analysis

The inventory analysis phase can be deemed the data collection step for the indicators defined under each subcategory (see section Supplementary Material for more details on indicators).

There are three types of indicators defined in this methodology:

• Basic requirement(s);
• Good social state;
• Bad social state.

A systematic procedure was followed in defining the above indicators. Our target was to propose one basic requirement and, at minimum, three indicators on each side (good or bad) per subcategory to fit into the 7-scale marking scheme described in the impact assessment section. Firstly, indicators of each subcategory were gathered from the literature; new indicators were proposed by abstracting the descriptions given in the SLCA guidelines to encapsulate more aspects within the subcategory of interest, where necessary. Secondly, those were classified into the basic requirement, good, and bad groups based on industry standards (e.g., ILO guidelines [34], UNEP/SETAC methodological sheets for subcategories in SLCA 2021 [6], and other relevant literature). Thirdly, a screening process was performed, where the indicators with similar meanings were merged into a single indicator having the same meaning. Owing to the effort put in at the initial stage, at least one indicator per group could be achieved. Fourthly, in such cases, existing indicators were cloned using a frequency or percentage factor; for instance, in proposing good indicators for “creation of local employment”, the percentage of the local workforce was used as the percentage factor. This percentage factor was extracted from one of our previous works, Dunuwila et al. [11]. Finally, all the indicators were rigorously reviewed for clarity by authors leveraging their expertise in industry and academia (see Table S1 of Supplementary Material) and an external panel of experts (see Table S2 of Supplementary Material). As a result, more than 300 indicators representing 39 subcategories could be proposed. However, we refrained from proposing indicators for the subcategory “smallholders including farmers” under “workers”, as it can be used in sector-level assessments [6]. The list of subcategories and corresponding indicators, including the supported literature and data sources, is provided in the tables under the Sections 1–6 in the Supplementary Material.

2.3.3. Impact Assessment

The role of this step is to evaluate the social state or performance at every subcategory listed in the UNEP/SETAC methodological sheets for subcategories in SLCA 2021 [6] using a novel social performance index (SPI). As mentioned earlier, impact assessment herein falls under Type-1, as this methodology uses basic requirements as a benchmark termed performance reference points to evaluate social performance [36].

• Development of Social Performance Index

The novel Social Performance Index was developed to measure social performance (at the subcategory level) across a product’s supply chain. We set a novel unit for SPI as Social Performance Hours (referred to as SPh). SPI can be defined as follows, with “working hours” and “social performance levels” (see Equation (1)) for any subcategory. The SPI can vary from −∞ (across the bad domain) to +∞ (across the good domain), and the larger the SPI, the better, and vice versa. We define two SPIs to distinguish the good and bad social states, i.e., ${SPI}_{k_{good}}$ (see Equation (2)) and ${SPI}_{k_{bad}}$ (see Equation (3)). The former sums up all good SPIs, while the latter sums up all bad SPIs across the subjected supply chain at subcategory k. We avoid referring to an average SPI value because good and bad effects can be neutralized during the averaging process.

$$Social performance index \left(SPI\right)=Social performance levels\times Working hours$$

(1)

$${SPI}_{k_{good}}={\sum }_{i=1}^n{SPI}_{k_{good, i}}$$

(2)

$${SPI}_{k_{bad}}={\sum }_{i=1}^n{SPI}_{k_{bad, i}}$$

(3)

where

${SPI}_{k_{good}}$: summation of good SPIs along the subjected supply chain at subcategory k;

${SPI}_{k_{good, i }}$: good SPI of the ith actor of the subjected supply chain at subcategory k;

${SPI}_{k_{bad}}$: summation of bad SPIs along the subjected supply chain at subcategory k;

${SPI}_{k_{bad, i}}$: bad SPI of the ith actor of the subjected supply chain at subcategory k;

$n$: total number of actors along the subjected supply chain.

We use working hours as an activity variable [3], which enables applying the SPI to the whole supply chain of a product/service. Activity variable is a measure of process activity with respect to process output [3]; hence, reflects the share of given activity associated with each unit process. Working hours at each process are calculated by a working hour model described in Pucciarelli et al. [37] (see Equation (4)). Data for the model can be extracted through field interviews, sustainability reports, other organization-specific reports, and sources such as SLCA databases [37]. This approach is premised on the understanding that social impacts are often generated during the production process, where labor input occurs. The social performance level component within our framework directly addresses the effects on stakeholders by evaluating the social impacts associated with those working hours.

$$Working hours per unit={\displaystyle \frac{W\times H\times n}{p}}$$

(4)

where

W: Number of workers involved in the process;

H: Worked hours per week;

n: Working weeks per year;

p: Production in the particular year.

Social performance levels are evaluated using a novel decision tree (see Figure 2). The primary role of this decision tree is to assess the subcategories with respect to the performances of indicators. The decision tree possesses seven levels: best, better, fair, basic, poor, worse, and worst. Each level is assigned a marking scheme of 3, 2, 1, 0, −1, −2, and −3 points, respectively.

• Basic level: This level is reached if all indicators defining basic requirements are satisfied and all good indicators are at “no” under a certain subcategory, or in the case of creation of local employment, contribution to economic development, and technology development (i.e., subcategories possessing positive notions), only if basic requirements are satisfied.
• Best level: This level is reached if all basic requirements are met and all good indicators are with “yes” under a certain subcategory, or in the case of the creation of local employment, contribution to economic development, and technology development if basic requirements are not satisfied and all good indicators are with “yes”.
• Better level: This level is reached if basic requirements are met and the good indicators with yes are larger than those with “no”, or in the case of creation of local employment, contribution to economic development, and technology development if basic requirements are not satisfied and the good indicators with “yes” are larger than those with “no”.
• Fair level: This level is reached if basic requirements are achieved and when good indicators with “yes” are equal to or smaller than those at “no”, or in the case of creation of local employment, contribution to economic development, and technology development if basic requirements are not satisfied and when good indicators with “yes” are equal to or smaller than those at “no”.
• Poor level: This level is reached when bad indicators having “yes” are equal to or smaller than those with “no”, or in the case of creation of local employment, contribution to economic development, and technology development if basic requirements are not satisfied and when bad indicators having “yes” are equal to or smaller than those with “no”.
• Worse level: this level is reached when bad indicators having “yes” are larger than those with “no”, or in the case of creation of local employment, contribution to economic development, and technology development if basic requirements are not satisfied and when bad indicators having “yes” are larger than those with “no”.
• Worst level: This level is reached when all bad indicators are at “yes”, or in the case of creation of local employment, contribution to economic development, and technology development if basic requirements are not satisfied and when all bad indicators are at “yes”.

The functionality of the decision tree can be exemplified as follows.

2.3.4. Interpretation

This step remains common to all other steps and interprets the outcomes of them being interactive with them.

2.4. Subcategories and Their Indicators in the Assessment

2.4.1. Workers

Ten subcategories (i.e., freedom of association and collective bargaining, fair salary, child labor, hours of work, forced labor, discrimination, health and safety, social benefits/security, employment relationships, and sexual harassment) defined in Traverso et al. [6] were considered for the methodology. However, we have refrained from proposing indicators for the subcategory “smallholders including farmers”, as we identified it was more suitable for sector-level analyses rather than site-specific analyses.

Basic requirement indicators for the subcategories are extracted from Traverso et al. [6], Ramirez et al. [28], Hannouf and Assefa [29], Osorio-Tejada et al. [32], ILO [34], Goedkoop et al. [38], and OSHA [39]. Subcategories except social benefits/security possess multiple indicators for basic requirements. For instance, as for the child labor subcategory, three basic requirements refer to ILO Convention No. 138 [34]: Is the organization in compliance with all child labor laws (national and international)? Is child labor avoided in activities that are not permitted? Are records detailing the names and ages or dates of birth of all employees maintained on file with strict adherence to compliance regulations? ILO Convention No. 138 refers to child labor as employing children less than 15 and 14 years of age in developed and developing or less developed countries, respectively. However, the same convention allows light work for underaged children without interfering with their education, while convention No. 182 [34] prohibits hazardous activities until 18 years of age. Here, we argue that a child employed within the legal limits can receive an additional income for his or her needs; hence, “Is child labor used in permitted activities with proper payment?”, and “Are scholarships or in-kind donations granted to permitted child workers?” are in place as good indicators. “Does the organization verify that employed children within the compulsory schooling age are attending school? (school attendees are supported; if non-attendees are observed, they are directed to schooling)” is in place to represent that working in the entity ensures the schooling of the child laborer for a better future.

Meanwhile, the indicator “Has the organization raised awareness of issues associated with child labor”? stands for an additional effort to educate the public on child labor. “Has the organization not recruited child laborers who are 14–18 years old?, Does the organization audit the rejections of actors who carry out unpermitted child labor in the upstream segment of the supply chain?, Does the organization audit the rejections of actors who carry out unpermitted child labor in the downstream segment of the supply chain?” are other good indicators for proactive involvement of the organization in eradicating child labor from its supply chain.

Four bad indicators are proposed based on ILO No. 138: Are there instances of child labor involved in prohibited activities? Is child labor used during the night? Are child workers punished for mistakes? Are child workers employed for restricted or unpermitted hours? We recommend avoiding interviewing management as a potential data source for bad indicators where possible, as it can result in biased answers.

The only basic requirement for the social benefits was based on Ramirez et al. [28]. Due to the lack of bad indicators in the literature for the same subcategory, a counting factor was used in proposing them (see Section 1 of Supplementary Material for more details).

2.4.2. Local Community

Nine subcategories (delocalization and migration, access to material resources, access to immaterial resources, local employment, cultural heritage, respect for indigenous rights, secure living conditions, community engagement, and safe and healthy living conditions) have been considered for the methodology. Their basic requirements were based on the UNEP/SETAC methodological sheets for subcategories in SLCA 2021 [6], Goedkoop et al. [38], Ramirez et al. [28], Osorio-Tejada et al. [32], and Harmens et al. (2022) [40]. Out of nine, six subcategories possessed multiple basic requirements. Due to the lack of good and bad indicators in the literature, most of them were newly proposed, abstracting UNEP/SETAC methodical sheets for subcategories in SLCA 2021 [6] or cloned using a percentage factor.

For example, community engagement has two basic requirements deriving from Traverso et al. [6], Dunuwila et al. [11], and Osorio-Tejada et al. [32]: Does the organization have effective interactions with the community (key informants) to avoid any harmful effects (neighborhood leaders, clergy, members of local authorities such as the municipal council, education providers, tourists and visitors, and business owners)? Has the organization implemented an effective mechanism to address queries and grievances of local communities? However, good indicators are derived from a set of percentage-based indicators in Dunuwila et al. [11] that represent the degree of effort the organization devotes to social welfare; for example, “Has the organization had evidence of supporting social welfare (10–5% of the profit has been allocated for community projects)?”. The only source of bad indicators for this subcategory was Osorio-Tejada et al. [32]. Bad indicators represent aspects where effective interactions with local community members are lacking, such as “Are access or guided visits of people from the community to the organization not allowed?” Please see Section S2 of the Supplementary Material for more information.

2.4.3. Society

The methodology includes seven subcategories (public commitment to sustainability issues, prevention and mitigation of conflicts, contribution to economic development, corruption, technology development, ethical treatment of animals, and poverty alleviation). Their basic requirements are from UNEP/SETAC methodological sheets for subcategories in SLCA 2021 [6] and Osorio-Tejada et al. [32]. Two out of seven subcategories have multiple basic requirements.

For instance, contribution to economic development possesses a positive notion; therefore, basic requirements represent the scenario where virtually no business is operating. Notably, basic requirements under “Contribution to economic development” (i.e., “Has the organization not reported positive or negative net income?”, “Has the organization had no research supporting economic development?”, and “Has the organization had no job creation or loss?”) are newly derived abstracting from Traverso et al. [6]. Good indicators tend to touch on aspects such as positive net job creation and positive net income and the commitment to economic development via R&D. The first and third are newly proposed indicators based on Traverso et al. [6], whereas the second is derived from Traverso et al. [6] and Osorio-Tejada et al. [32]. All bad indicators were newly proposed, referring to Traverso et al. [6], and denoted opposite meanings to the indicators on the good side (negative net income, negative net job creation, and a falsified approach in economic research). Please refer to Section S3 of the Supplementary Material for more information on indicators under this subcategory.

2.4.4. Children

Three subcategories—education provided in the local community, health issues for children as consumers, and children’s concerns regarding marketing practices—were considered for the methodology. As a newly defined stakeholder category in UNEP/SETAC methodological sheets for subcategories in SLCA 2021 [6], all indicators of the aforementioned subcategories were extracted from that document. Only one subcategory had multiple basic requirements, while the others only had one. For example, “Health issues for children as consumers” had one basic requirement: “Have health problems not been reported?” Good indicators were based on aspects of organizational commitment to improving the physical, mental, and social health of the children; for example, “Has the organization carried out programs to improve the social health of children? Opposite connotations of the good indicators were considered for proposing bad indicators; for instance, “Has the organization carried out activities that harm the mental health of children?” Please refer to Section S4 of the Supplementary Material for more information on the indicators of this subcategory.

2.4.5. Value Chain Actors

The methodology has considered five subcategories: fair competition, supplier relationships, respect for intellectual property rights, promoting social responsibility, and wealth distribution. The basic requirements of these subcategories were based on UNEP/SETAC methodological sheets for subcategories in SLCA 2021 [6], Ramirez et al. [28], and Hannouf and Assefa [29]. Four subcategories had multiple basic requirements. For example, the basic requirements of wealth distribution are “Are contractual instruments present that ensure the distribution of value among the actors within the supply chain?” and “Does the organization assure a fair price to protect all value chain actors?” Three newly proposed good indicators are evident. One describes the organization’s assistance extended to troubled value chain actors, while others represent the organizational effort to disseminate fair pricing across the value chain. Bad indicators, which are newly proposed, represent the opposite of basic requirements or the consequences of their absence, as well as the organization’s involvement in demoting fair pricing across the supply chain. Please refer to Section S5 of the Supplementary Material for additional information on indicators.

2.4.6. Consumers

The methodology includes five subcategories: health and safety of consumers, feedback mechanism, consumer privacy, transparency in social and environmental issues, and end-of-life responsibility. The basic requirements of these subcategories were based on UNEP/SETAC methodological sheets for subcategories in SLCA 2021 [6], ISO [41], GRI [42], Osorio-Tejada et al. [32], and Ramirez et al. [28]. Only “Health and safety of customers” had multiple basic requirements, which are “Are management measures present to assure consumer health and safety, such as labels, consumer surveys, etc.?” and “Has the organization had no complaints/evidence of health and safety issues from consumers?” Good indicators of the same included perspectives such as funding consumers (of other companies) to recover from health and safety issues, educational programs conducted to improve the health and safety of consumers, and whether the product of interest improves the health and safety of consumers. Except for the indicator representing the last aspect, all the indicators were newly proposed per the UNEP/SETAC methodological sheets for subcategories in SLCA 2021 [6]. On the other hand, bad indicators of the same subcategory were based on indicators describing evidence of customer health and safety issues and the absence of management measures to ensure consumer health and safety. See Section S6 of Supplementary Material for more details on these indicators.

2.5. Data Collection

In order to evaluate defined indicators, site-specific data from the past three years is required. Potential data sources for each indicator are mentioned in the Supplementary Material. There are three categories of data sources: primary, secondary, and tertiary. Primary data sources include site audits, interviews with individuals (both inside and outside the organization), and questionnaire surveys, while secondary data sources include documents directly from the organization of interest, such as sustainability reports. Tertiary data sources are the sources that are indirectly related to the organization or prepared by a third party, such as reports from non-governmental organizations (NGOs). We suggest gathering as much information as possible from primary sources to make the evaluation more credible. If impossible, one can rely on secondary or tertiary data sources. The exemplary layout in

Table 1. Evaluation of the subcategory of freedom of association and collective bargaining at the rubber-tapping stage of the RSS manufacture in the audited rubber estate. BR, G, and B, N/A refer to basic requirements, and good and bad indicators, not applicable, respectively.

Stakeholder	Subcategory	Indicator	Status	Yes/No	Remarks	Data Source (Please Mention here If data Triangulation is Performed)
Workers	Freedom of association and collective bargaining	Is the downstream segment of the supply chain audited for rejections of actors who cause freedom of association and collective bargaining issues?	G	No	No audits are carried out for the downstream supply chain	Field interviews with management.
		Is the upstream segment of the supply chain audited for rejections of actors who cause freedom of association and collective bargaining issues?	G	N/A	Since the upstream plantations are also owned by the factory this indicator automatically becomes N/A.	Field interviews with management.
		Can unions/collective activities influence the organization to improve working conditions?	G	No	No specific influence on improving working conditions.	Field interviews with workers (verified interviewing workers’ union)
		Does the company allow union meetings/collective activity meetings on the company premises?	G	No	The union meetings are not held on factory premises.	Field interviews with workers (verified interviewing workers’ union)
		Is there a transparent scheme available to grant duty leave for union meetings/collective activities?	G	No	Special leaves are not granted.	Field interviews with workers (verified interviewing workers’ union)
		Are employee(s)/union representatives invited to contribute to the planning of more significant changes in the organization that will affect the working conditions?	G	No	Union representatives are not invited to key meetings that affect the working conditions.	Field interviews with workers (verified interviewing workers’ union)
		Do the workers have a proper mechanism to promote, defend and negotiate their respective interests, collectively and freely? (At least one of the activities listed below is present and/or workers are free to form them) (e.g., Unions and other collective activities)	BR	Yes	Unions are present and workers are free to form them.	Field interviews with workers (verified interviewing workers’ union)
		Are workers free to join the above activities of their choosing?	BR	Yes	Workers are free to join the unions of their choosing.	Field interviews with workers (verified interviewing workers’ union)
		Are there no incidents reported to prevent the above activities?	BR	Yes	No such incidents were reported.	Field interviews with workers (verified interviewing workers’ union and management)
		Are unions/collective activities absent or are workers not free to form/request them?	B	N/A	N/A	N/A
		Do workers have no freedom to join unions/collective activities of their choosing?	B	N/A	N/A	N/A
		Are there any incidents reported on the prevention of the above collective activities?	B	N/A	N/A	N/A

can be useful for a transparent indicator evaluation. A pedigree matrix can be used to indicate data uncertainty if necessary [43]. An example is provided in Supplementary Material (see Table S3) to assess the four aspects of data uncertainty to preserve the transparency and traceability of data: reliability of the source(s), temporal conformance, geographical conformance; and technical conformance.

Data triangulation (confirmation of data using multiple actors) has been deemed an important practice in LCA; therefore, it is recommended to perform data triangulation when necessary [28]. For instance, if a specific indicator under workers was evaluated by interviewing workers, this information can be confirmed by interviewing the human resources department and workers’ union or by reviewing secondary data sources such as the organization’s sustainability reports.

3. Discussion and Outlook

The following case study on the manufacture of Ribbed Smoked Sheets (RSS) elaborates on the functionality of the proposed methodology. RSS is an important primary rubber product for many value-added products like tires and shoes. Figure S1 depicts the processes associated with RSS manufacture and the system boundary, defined as cradle-to-gate, considered for the case study.

As per Figure S1, the manufacture of RSS begins with the manual tapping of rubber trees at rubber plantations, after which the rubber latex is taken to factories [44]. Upon arrival, the latex is poured into flat pans, diluted with water, and treated with formic acid to perform coagulation. The solid coagulum is then milled using manually operated smooth and grooved rollers. The resulting sheets are rinsed, hung to drip-dry, and then smoke-dried for three to five days. Finally, the sheets are weighed and sent to distribution centers.

Data required for marking the indicators (for example, see Table 1) and working hours were extracted from visiting a rubber estate that manufactures RSS in Sri Lanka and interviewing relevant stakeholders. Data collected were from within the past 3 years. Several subcategories and indicators had to be excluded due to their applicability identified during the interviews (see Table 1 and Table S10 for examples).

Table 1 summarizes the evaluation of the subcategory “Freedom of association and collective bargaining” at the rubber-tapping stage of the RSS manufacture in the audited rubber estate. As per Table 1, the three basic requirements have been fulfilled; hence, moving to the good domain is possible. Then at that point, “are all good indicators “Yes”?”. The path dedicated to “No” shall be taken as no good indicators out of five are with “Yes”. Again, the path “Yes” is to be taken at “Are all good indicators “No”?”. Here, we head ultimately to the social performance level of “Basic” (see the caption of Table S4 for another indicator evaluation).

The results could are summarized in Figure 3 and Figures S2–S6. For clarity, a sample calculation for the subcategory “Social benefits” is provided, and the information required is given in Figure 4. Referring to the resulting SPIs of all subcategories (see Figure 3 and Figures S2–S6), the most influential hotspot (affected subcategory) that is common to both tapping and processing was observed to be with social benefits of the audited estate. The baseline scenario (current scenario), representing the current situation, can be calculated as in Equation (5). For the betterment of that subcategory, if we assume that the organization provides retirement, disability, and survivors’ benefits; paid maternity and sick leave; and education and training to the workers, the fulfillment of basic requirements occurs. Therefore, the degree of improvement can be calculated following Equation (6). For the sake of simplicity in calculations, social impacts from the background processes, such as chemical production and electricity generation, were neglected.

The baseline scenario, as per Figure 4, is written as follows:

$$\begin{gathered} {SPI}_{{Social Benefits}_{bad}}=-3\times 1+-3\times 173\times 1=-900+(-519)=-1419 \mathrm{SPh}/\mathrm{tonne} \mathrm{of} \mathrm{RSS} (\mathrm{social} \mathrm{performance} \mathrm{level} \\ (\mathrm{no} \mathrm{unit})\times \mathrm{working} \mathrm{hours} \mathrm{per} \mathrm{unit} \mathrm{weight} (\mathrm{h}/\mathrm{tonne})\times \mathrm{quantity} (\mathrm{tonne}\left)\right) \end{gathered}$$

(5)

The improved scenario, as per Figure 4, is written as follows:

$${SPI}_{{Social Benefits}_{good/bad}} = 3 \times 300 \times 1 + 0 \times 173 \times 1 = 0 \mathrm{SPh}/\mathrm{tonne} \mathrm{of} \mathrm{RSS}$$

(6)

As a result of the improvement, ${{SPI}^{\prime }}_{{Social Benefits}_{bad}}$ has become zero from the SPI of −1419 SPh; overall, this change can be deemed a 100% less-bad movement of impacts.

Figure 5 illustrates fictional results that can emerge after applying our methodology to a product supply chain. The subcategories listed in the figure are related to the workers’ stakeholder group (for ease of indication here, we only report the impacts of workers). The perforated line represents the altered good (green zone) and bad impacts (red zone) after the insertion of an improvement option or consideration of a scenario. The black, yellow, orange, and purple arrows represent the more-good, less-bad, less-good, and more-bad impacts, respectively. To facilitate comparisons with other options or scenarios, the changes in impacts within a specific subcategory may be reported along with the degree of change (for example, in the case of freedom of association and collective bargaining, less-bad: −20% and more-good: 5%). Improvement options may be proposed after observing the hotspots along the supply chain (detecting processes with higher negative SPI values using a contribution analysis [45]), based on predefined scenarios, or at the discretion of the LCA practitioner. In some instances, certain stakeholders, subcategories, and indicators may not be applicable to certain factories or organizations. In such cases, we recommend omitting them in the order of stakeholder category, subcategory, and indicator. Aforesaid applicability can be assessed at the discretion of organizational or factory personnel or LCA practitioners.

The frameworks that are closest to our methodology are product social impact assessment [30] and the frameworks in Osorio-Tejada et al. [32] and Corona et al. [31]. However, none encompasses every subcategory in the most recent UNEP/SETAC methodological sheets for SLCA [6] while resting on the grey term “positive impacts”. Our framework selects performance levels using a decision tree, whereas the product social impact assessment and Osorio-Tejada et al. [32] use predefined indicators for observing performance levels. We believe our methodology is more flexible and adaptive than those because the indicators can be modified or the number of them can be increased at the discretion of the LCA practitioner (N.B., we believe more indicators can increase the credibility of the analysis). Though Corona et al. [31] used a decision tree, its role has been constricted to evaluating each indicator using the national average as the reference level; therefore, it cannot be applied universally (N.B., the task of our decision tree was to evaluate subcategories based on universally applicable basic requirements, good and bad indicators). Finally, therein, the evaluated indicators are averaged to obtain subcategory scores, which allows for effect cancellation (i.e., good effects cancel bad effects). In addition, none of the aforementioned studies could link the social impacts to a functional unit. However, with our methodology, this gap has been filled, and the impacts can be quantified with reference to a functional unit (see Figure 3, which summarizes the social impacts in SPh per tonne). Considering what has been stated previously, our methodology will make it possible to conduct more versatile quantitative social analyses.

ISO 14075 provides a standardized framework for social life cycle assessment, emphasizing a comprehensive understanding of both positive and negative social impacts (however, it offers limited specific guidance on assessing positive social impacts) [4]. Our framework complements this by offering a method for quantifying the social impacts associated with specific activities, explicitly differentiating between ‘more-good’ and ‘less-bad’ outcomes. This approach directly addresses ISO’s call for comprehensive impact assessment, enabling a more detailed understanding of the overall social implications of a product or service and encouraging organizations to actively seek opportunities to improve social well-being.

In addition, we believe that the social handprint calculated by our methodology offers valuable comparability because the information used (i.e., more-good and less-bad impacts) is evaluated against universally applicable basic requirements, rather than the context-dependent ‘business as usual’ approach often found in conventional social handprints [25]. This allows for a more standardized comparison of the same product across different supply chains or different products. Further, conventional handprints often address social risks, while our new handprint focuses on the spectrum of social impacts, incorporating both positive and negative changes of social impacts or more-good and less-bad social impacts (see the example given referring to Figure 4). We propose referring to this refined metric as the “social headprint”, stressing its grounding in fundamental needs and its potential for broader application. However, further research is necessary to empirically validate its functionality and compare its outcomes with other handprint methodologies.

Supply chains can stretch across multiple countries. As a result, extracting site-specific data can be challenging. Such instances can be averted using social LCA databases (e.g., PSILCA [46] and the Social Hotspot Database [43]). However, the so-called good domain remains undiscovered, as these databases report social risks rather than social performance. That is, either “less-good” or “more-bad” movements can be detected. Also, a limited number of subcategories can be evaluated using these databases (for instance, currently the PSILCA database [46] covers only 19 subcategories). To resolve this issue to some extent, using a relational correspondence matrix where social performance levels are assigned to social risk levels or developing a database that evaluates the “potential good effects of products” supply chains [5] can be given. However, the former looks more straightforward than the latter.

Basic requirements can change over time as the current labor and other industry standards are prone to change; therefore, adjustments to such may be required as time passes. A further limitation lies in the potential for cultural variability to influence the definition and interpretation of “basic requirements”. As labor laws, social norms, and values differ across regions, the perceived adequacy of existing labor practices may vary significantly. While our case study offers a specific example within Sri Lanka, future applications may prioritize context-specific indicator definitions and flexible thresholds, incorporating input from local stakeholders to reduce culturally biased assumptions. This stakeholder engagement involves actively seeking perspectives from workers, employers, and community members to understand their views on fair labor practices within their specific socio-cultural environment. To enhance the methodology’s adaptability, indicators can be adjusted based on the cultural and socioeconomic context of the study area. For instance, what forms an acceptable level of workplace safety or a fair wage may be understood differently depending on the cultural context. Therefore, researchers and practitioners must be mindful of these distinctions when applying the methodology in diverse settings, ensuring that the assessment reflects the specific values and priorities of the local stakeholders. This also demands a detailed contextual analysis, involving examination of local labor laws, social norms, cultural values, and economic conditions to identify any potential conflicts or distinctions that may affect the interpretation or implementation of “basic requirements”.

Further, a way to aggregate the good and bad impacts into stakeholder and impact categories may be necessary to smoothen the interpretation and decision-making processes. In this regard, good or bad effects must not be neutralized. Another improvement would be weighing the indicators; for instance, in our methodology, all the indicators are considered to have equal weightage. In a real situation, an indicator like “Are there fatal accidents in the organization? (no fatal accidents desired)” should possess a higher importance level than “Are there serious nonfatal accidents in the organization? (no serious non-fatal accidents desired)”. A technique like Analytic Hierarchical Process (AHP) can be useful in this regard [47].

When applying the SPI framework, it is important to recognize that a single activity can simultaneously generate good and bad social impacts across various stakeholders. For example, a new factory might create jobs while also increasing noise and straining local infrastructure. Such mixed impacts need to be carefully detected and quantified, considering the perspectives of those affected. Developing a trade-off index, such as the trade-off valuation index [48], is recommended for future research, along with carefully assessing the cause–effect mechanisms to achieve a holistic understanding.

We acknowledge that using qualitative data and expert judgment introduces some subjectivity when we decide on the ‘yes/no’ values for our indicators. To address this, several steps can be taken to make our process more consistent and transparent. These steps included using a standardized interview guide, having multiple researchers involved in analyzing the data, and setting up clear rules for how to interpret the information we gathered. We recognize that a fully standardized scoring system would be ideal. However, because our indicators are very specific to the context we are studying, a data-driven approach to setting the ‘yes/no’ thresholds may be more appropriate. This means that the thresholds were determined by analyzing the data we collected, rather than relying on pre-set benchmarks.

We are to apply this methodology to different supply chains as the next step.