Int. J.

Production Economics 270 (2024) 109178

Contents lists available at ScienceDirect

International Journal of Production Economics

journal homepage: www.elsevier.com/locate/ijpe

Data-driven digital transformation for supply chain carbon neutrality:

Insights from cross-sector supply chain
Amine Belhadi a, Mani Venkatesh b, *, Sachin Kamble c, Mohammad Zoynul Abedin d
a
Rabat Business School, International University of Rabat, Morocco
b
Montpellier Business School, France, and Asper School of Business, University of Manitoba, Winnipeg, Canada
c
EDHEC Business School, Roubaix, France
d
School of Management, Swansea University, Bay Campus, Fabian Way SA1 8EN Swansea, UK

A R T I C L E I N F O A B S T R A C T

Keywords: Following the growing pressure on firms and supply chains regarding their environmental impact, carbon
Data-driven digital transformation neutrality of supply chains is gaining substantial attention among scholars and practitioners. Data-driven digital
Data-analytics capabilities transformation supports supply chains in achieving higher carbon reduction while improving efficiency and
carbon transparency
economic performance. However, the conditions under which data-driven digital transformation can provide the
carbon uncertainties
carbon neutrality
desired effect remain unclear due to a lack of empirical evidence. This study aims to address this gap by
OIPT examining how data-driven digital transformation, enabled by data analytics capabilities, contributes to estab­
lishing a win-win situation between carbon and economic performance in the face of several sources of carbon
uncertainty through fostering supply chain carbon transparency. Drawing upon the organizational information-
processing theory, we posit that the fit between information needs to reduce carbon uncertainties and the in­
formation capabilities provided by data-driven digital transformation is critical for enhancing supply chain
carbon transparency and balancing supply chains’ economic and carbon performance. We examine these re­
lationships using regression tests based on survey data from 437 manufacturing companies from different regions
(i.e., Europe, Africa, and Asia). Our results reveal that data analytics capabilities alone cannot enhance supply
chain carbon transparency until integrated into a comprehensive business transformation. In that case, carbon
transparency would positively mediate overcoming carbon uncertainties and improve the supply chains’ carbon
and economic performance.

1. Introduction standards (Lintukangas, et al., 2022). The drive towards carbon

neutrality is increasingly aligning with adopting data-driven digital
As climate change intensifies, evidenced by extreme weather events, technologies. These technologies, from big data analytics to Artificial
the critical role of supply chains in global carbon emissions has come Intelligence, are redefining the approach to sustainable supply chain
into sharp focus. This has led to an increased emphasis on achieving management, moving away from traditional practices towards more
carbon neutrality in supply chains, recognizing their substantial impact efficient, technology-driven solutions (Sheng, et al., 2023). Recent
on environmental sustainability. Today’s businesses are not only ex­ trends showcase the synergy between data-driven digital transformation
pected to be transparent about their carbon footprints but are also urged and carbon neutrality, as evidenced by Walmart’s Gigaton program.
to actively pursue carbon neutrality in their operations and supply Utilizing data analytics, this initiative significantly reduces greenhouse
chains (Jira and Toffel, 2013; Villena and Dhanorkar, 2020; Qian and gas emissions by enhancing energy and transportation efficiency within
Schaltegger, 2017). This imperative reflects a shift from mere awareness its supply chain, akin to the impact of removing numerous vehicles from
to action, as supply chains are critically positioned at the intersection of the road (Walmart Sustainability Hub, 2020). This illustrates a broader
environmental impact and business operations. The transition to shift where businesses integrate data-driven digital transformation to
carbon-neutral supply chains is a strategic priority for sustainable achieve economic goals alongside carbon neutrality. The importance of
business practices and compliance with evolving global environmental examining digital transformation’s role in supply chain dynamics and

* Corresponding author.
E-mail addresses: [email protected], [email protected] (M. Venkatesh).

https://doi.org/10.1016/j.ijpe.2024.109178
Received 17 April 2023; Received in revised form 16 December 2023; Accepted 7 February 2024
Available online 20 February 2024
0925-5273/Crown Copyright © 2024 Published by Elsevier B.V. All rights reserved.
A. Belhadi et al. International Journal of Production Economics 270 (2024) 109178

carbon neutrality is thus highlighted. chain carbon transparency?

The pursuit of carbon neutrality through data-driven digital trans­
RQ2. How does supply chain carbon transparency, facilitated by data-
formation not only amplifies operational efficiency, customer value, and
driven digital transformation, influence the dual objectives of economic
overall performance but also integrates data analytics capabilities into
and carbon performance in the presence of carbon uncertainties?
strategic, structural, and procedural aspects of supply chain manage­
To address the research above questions, we posit that firms oper­
ment (Belhadi, et al., 2021; Enrique et al., 2022). However, the trans­
ating in uncertain environments need to process more information,
formation’s pace varies across firms, influenced by factors such as
which can be facilitated by the development of information processing
capital requirements and the risk of financial strain, particularly in in­
capacity (Srinivasan and Swink, 2018; Aben et al., 2021; Zhu et al.,
dustries like manufacturing, known for significant carbon emissions
2018). In this study, we argue that data-driven digital transformation
(Papanagnou, et al., 2022; Sheng et al., 2023; Villena and Dhanorkar,
represents a form of information-processing capacity that can help firms
2020). This variation in the pace of digital transformation across firms
overcome carbon uncertainties and enhance supply chain carbon
necessitates empirical research to explore its impact on supply chain
transparency, thus influencing economic and carbon performance. To
performance. We define ’supply chain carbon performance’ as the
test our argument, we conducted a quantitative survey of 437 firms from
effectiveness of a supply chain in managing and reducing its carbon
various regions (i.e., Europe, Africa, and Asia) belonging to different
emissions, a crucial aspect of environmental sustainability (Villena and
supply chains. The collected data were analyzed using regression and
Dhanorkar, 2020). Similarly, ’supply chain uncertainties’ refer to the
bootstrapping techniques.
varied challenges and unpredictabilities in supply chain operations,
Our research distinctly contributes to the literature on supply chain
especially regarding carbon management. These uncertainties can arise
management by providing empirical insights into the application of
from fluctuating market demands, supply chain disruptions, and
data-driven digital transformation for achieving carbon neutrality.
evolving environmental regulations (Sheng, et al., 2023).
Diverging from prior theoretical explorations, this study specifically
The imperative for harmonizing economic performance with carbon
examines the role of digital transformation in enhancing carbon trans­
reduction emphasizes the critical role of supply chain carbon trans­
parency within supply chains, particularly in contexts marked by un­
parency (Jira and Toffel, 2013; Hahn et al., 2015). This transparency,
certainty. The findings illuminate the intricate balance between
which involves providing stakeholders with detailed carbon emission
achieving carbon neutrality and maintaining economic performance,
information, is increasingly recognized as crucial for meeting sustain­
offering a nuanced understanding of how supply chains can navigate the
able development goals and making informed decisions to mitigate
dual challenges of environmental responsibility and business efficacy.
environmental impact while enhancing operational efficiency (Villena
and Dhanorkar, 2020; Hassan and Romilly, 2018; Liesen et al., 2017). In
2. Background literature
this context, supply chain carbon transparency emerges as a critical
mechanism through which data-driven digital transformation can
2.1. Organizational Information Processing Theory
harmonize economic performance and carbon neutrality. This align­
ment is supported by the Organizational Information Processing Theory
OIPT, as conceptualized by Galbraith (1974), provides a robust
(OIPT), which highlights the importance of effective information man­
framework for understanding organizational information processing,
agement in gaining a competitive edge. OIPT provides a theoretical lens
particularly in contexts of high uncertainty. This theory is instrumental
to understand how integrating data-driven digital transformation with
in explaining how firms can effectively manage information by reducing
supply chain strategies can align with broader organizational goals,
their need via “mechanistic” strategies or enhancing their processing
influencing sustainable supply chain strategies and overall performance
capabilities. The relevance of OIPT in supply chain management, a field
in a rapidly evolving business environment (Galbraith, 1974; Simon,
characterized by inherent uncertainties, is well-established (Srinivasan
1957). Consequently, data-driven digital transformation equips firms
and Swink, 2018). The advent of data-driven digital transformation has
with the necessary human, technical, and organizational capabilities to
extended OIPT’s applicability in supply chain contexts. Aben et al.
manage information effectively, particularly in supply chain carbon
(2021) explore how digital transformation necessitates and facilitates
transparency initiatives.
enhanced information processing capabilities in organizations. This is
A critical gap exists in current research on carbon transparency
particularly relevant in managing supply chain carbon transparency,
within supply chains, particularly its empirical validation in the era of
where the processing of extensive carbon data is crucial for operational
data-driven digital transformation. While theoretical models posit that
and environmental decision-making. In aligning economic performance
carbon transparency could be a critical mediator between digital
with carbon neutrality, the role of OIPT becomes increasingly signifi­
transformation and the dual goals of economic performance and carbon
cant. Galbraith’s (1974) proposition regarding information processing
neutrality (Lintukangas, et al., 2022; Shui et al., 2022; Sheng et al.,
aligns with firms’ challenges in balancing these twin objectives within
2023), empirical evidence to support these assertions is notably lacking.
their supply chains. Zhu et al. (2018) reinforce this perspective, high­
This gap is more pronounced considering the challenges in balancing
lighting the effectiveness of information processing strategies in navi­
economic efficiency with carbon footprint reduction in supply chain
gating the complexities of sustainable supply chain management.
operations. The complexity and uncertainty of supply chain carbon data
Empirical studies further validate OIPT’s utility in the era of data-
further accentuate the need for a robust, empirically grounded frame­
driven digital transformation. For example, research demonstrates that
work. Such a framework is essential to explore and validate the inter­
leveraging digital tools for information processing can significantly
twined relationship between economic performance and carbon
enhance decision-making in environmental sustainability (Srinivasan
management in supply chains, moving beyond theoretical conjectures to
and Swink, 2018). Such evidence underscores the theory’s effectiveness
practical, data-driven insights. Addressing this gap, our study contrib­
in understanding the dynamics of information management in digitally
utes novel empirical insights, setting itself apart from existing theoret­
transformed supply chains. Therefore, our study’s application of OIPT
ical discourse. We provide concrete, data-driven evidence of how digital
offers a theoretical basis to investigate how data-driven digital trans­
transformation facilitates carbon neutrality in supply chains, offering a
formation impacts supply chain carbon transparency. This approach is
practical framework for effectively understanding and implementing
consistent with OIPT’s principles and provides new insights into its
these technologies.
application in supply chain carbon management, particularly in the
This paper proposes an integrated conceptual framework to answer
digital age.
two fundamental research questions (RQs).
RQ1. How can data-driven digital transformation influence supply

2
A. Belhadi et al. International Journal of Production Economics 270 (2024) 109178

2.2. Data-driven digital transformation have emerged nationally, geared toward increasing transparency and
improving emissions management (Hahn, et al., 2015; Luo et al., 2018).
Digital transformation involves the integration of digital technolo­ Drawing upon the literature on transparency (as reviewed by Shui
gies into business operations, processes, and value creation, resulting in et al. (2022) and Schnackenberg et al. (2021)), supply chain carbon
a transformation of the organization (Sheng, et al., 2023; Papanagnou transparency can be defined as the provision of high-quality and specific
et al., 2022; Govindarajan, Srivastava & Mani, 2021). Hanelt et al. carbon emission data by and to all supply chain partners and stake­
(2021) defined digital transformation as the organizational trans­ holders (Villena and Dhanorkar, 2020). In this context, our conceptu­
formation induced and shaped by the pervasive dissemination of digital alization of supply chain carbon transparency encompasses three pivotal
technologies across the organization. Data-driven technologies such as dimensions: supply chain disclosure, information accuracy, and infor­
big data (Matthias, et al., 2017), data analytics (Akter, et al., 2016), mation clarity. Supply chain disclosure pertains to a supplier’s commit­
artificial intelligence (Fosso Wamba, et al., 2022), and machine learning ment to openly sharing relevant carbon information with all
(Filom, et al., 2022) are at the center of these technologies, using either stakeholders, ensuring transparency extends beyond selective disclosure
data analytics capabilities (Papanagnou, et al., 2022). According to to influential parties. Transparent suppliers, in contrast, embrace
Akter et al. (2016), data analytics capabilities enable firms to extract openness by sharing information even with stakeholders who might
business insights by combining three elements, i.e., technology, man­ critically assess such data, including non-governmental organizations
agement, and human skills. This has been demonstrated to be a pre­ (NGOs) and the wider public. Information accuracy underscores the
requisite capability to transform businesses into a competitive force essential reliability of carbon-related data provided by the supplier.
(Awan, et al., 2021; Mikalef et al., 2019). Based on the literature on Transparent information should be free from bias and grounded in
digital transformation (for a detailed review, see Hanelt et al., 2021; robust data sources to ensure its credibility. Information clarity empha­
Gong and Ribiere, 2021), we define "data-driven digital transformation" sizes the extent to which suppliers communicate environmental infor­
as a fundamental process triggered by the innovative use of data ana­ mation. Greater clarity and comprehensiveness of carbon information
lytics capabilities to radically improve organizational performance. This contribute to higher transparency. Thus, these three dimensions
transformation is powered by the vast data generated within the firm’s collectively shape the overall degree of supplier transparency, high­
environment, often called "the new oil" (Aben, et al., 2021), driving lighting that each dimension uniquely establishes a comprehensive
transformative changes facilitated by the firm’s ability to process in­ transparency framework. While these definitions may not be standard­
formation effectively. From an OIPT perspective, data-driven digital ized in the strictest sense, we have synthesized insights from established
transformation empowers firms to enhance their capacity for processing literature sources to ensure their relevance and coherence within our
information. They can proactively gather data from various facets of study’s scope.
their operational landscape and strategically process it to instigate and According to Shui et al. (2022), although the outcomes resulting
steer profound organizational change (Srinivasan and Swink, 2018). from supply chain carbon transparency in carbon neutrality are ex­
This definition underscores the synergy between data analytics capa­ pected to attract the most scholarly attention, this is not the case in the
bilities and information processing capacity, collectively driving the current literature. Even worse, the few available investigations of the
metamorphosis brought about by digital transformation. We follow association between carbon transparency and carbon neutrality reveal
several information systems studies to reflect on data analytics capa­ conflicting results. Based on the relational view, the study by Lintu­
bilities (Srinivasan and Swink, 2018; Akter et al., 2016) and digital kangas et al. (2022) highlighted the role of carbon information trans­
transformation (Belhadi, et al., 2021) as the intersection of three ele­ parency in achieving the supply chain carbon management goals. Qian
ments, i.e., infrastructure, processes, and skills. Infrastructure refers to and Schaltegger (2017) propose that supply chain carbon disclosure can
the IT tools and digital technologies that drive data-driven digital be a valuable management tool for companies seeking to enhance their
transformation. This includes data collection tools (e.g., sensors, big carbon performance by reducing their emission intensity. Their research
data …) and data-processing technologies (e.g., machine learning, indicates that businesses can subsequently achieve a positive correlation
artificial intelligence …). Processes are defined as organizational struc­ with supply chain carbon neutrality by improving the quality of infor­
tures and instances that direct infrastructure use toward the firm’s mation disclosed. However, other studies, such as Hassan and Romilly
maximum value. Skills refer to the human competencies and abilities (2018) and Kim and Lyon (2011), found no causal link between carbon
needed to leverage the insights drawn from data analytics processes to transparency and carbon performance. They even discovered that some
drive organizational transformation. Therefore, data-driven digital companies used participation in disclosure programs to greenwash their
transformation is a function of these three dimensions, each contributing emissions. Their research indicated that disclosure programs had no
uniquely to the change achieved through data. significant impact on changes in carbon performance. Despite the con­
flicting findings in the past literature, the question, according to
2.3. Supply chain carbon transparency and carbon performance Schnackenberg et al. (2021), needs to go beyond positive/negative as­
sociation to the underlying mechanisms and conditions under which
In our study, ’supply chain carbon performance’ is expressly defined carbon neutrality can be an outcome of carbon transparency in the
to capture the effectiveness of a supply chain in reducing its carbon context of the supply chain.
emissions. This definition aligns with the understanding of contempo­
rary supply chain management literature, where reducing carbon 3. Theoretical framework and hypotheses
emissions is increasingly seen as a critical aspect of operational effi­
ciency and sustainability (Villena and Dhanorkar, 2020; Hassan and Grounded on the premises of the OIPT, we propose the theoretical
Romilly, 2018). ’Carbon performance’ in this context refers to the research framework depicted in Fig. 1. The framework focuses on the
quantification of emissions and the strategies and actions taken to organizational fit between data-driven digital transformation as the
minimize these emissions across the supply chain. information-processing capacity and supply chain carbon transparency
In today’s focus on carbon neutrality, increasing carbon trans­ as the information-processing need. Afterward, we investigate the
parency in the supply chain has become a growing interest for com­ spillover effect on both carbon neutrality and economic performance.
panies due to several regulatory and operational purposes (Villena and Accordingly, we compare the organizational fit among the elements of
Dhanorkar, 2020). On an international level, there has been a rise in data-driven digital transformation, i.e., data analytics capabilities (H1),
voluntary initiatives, such as the nonprofit Carbon Disclosure Project business transformation (H2), and their combination (H3). Additionally,
(CDP), compelling companies to disclose their greenhouse gas (GHG) our model contends that supply chain carbon transparency affects both
emission efforts and performance. Additionally, mandatory initiatives carbon neutrality and the economic performance of the supply chain and

3
A. Belhadi et al. International Journal of Production Economics 270 (2024) 109178

Fig. 1. Theoretical framework.

acts as a mediator between data-driven digital transformation and visibility and transparency, such as agility and collaboration. Thus, we
supply chain carbon neutrality (H4) and economic performance (H5). can project these premises to supply chain carbon transparency to hy­
Lastly, we consider supply chain uncertainties as information-boundary pothesize that.
conditions that create tensions for the information-processing fit (H6).
H2. Business transformation leads to higher supply chain carbon
After that, we detail the suggested hypotheses illustrated in Fig. 3,
transparency.
associating data-driven digital transformation, uncertainties, and car­
Given the evidence above, combining data analytics capabilities and
bon transparency with two-level performance, i.e., carbon neutrality
business transformation, i.e., data-driven digital transformation, is
and economic performance.
deemed to have mutually reinforcing effects on supply chain carbon
transparency. The reason behind this assumption is the large body of
3.1. Data-driven digital transformation and supply chain carbon knowledge asserting that data-driven technologies constitute an enabler
transparency for business transformation toward more transparency in several con­
texts (Keller, et al., 2021; Pappas, 2018). Hence, we assume that.
As discussed earlier, data-driven digital transformation can be
defined through two dimensions, i.e., data analytics capabilities and H3. Data-driven digital transformation (the combination of the two
business transformation. Regarding data analytics capabilities, past dimensions, i.e., data analytics and business transformation) has a
literature has primarily established a positive link between data-driven greater effect on increasing supply chain carbon transparency than only
technologies and transparency in the supply chain context. For instance, one of the two dimensions.
Shafiq et al. (2020) and Fernando et al. (2018) established that supply
chain analytics have significant spillover effects on supply chain trans­
3.2. The mediating effect of supply chain carbon transparency
parency based on a resource-based perspective. The same finding has
been reported in the study of Zhu et al. (2018) from an OIPT point of
OIPT asserts that the fit between information and its proper utiliza­
view. Indeed, the OIPT asserts that a result of developing data analytics
tion is a potential source of enhanced performance at different levels. In
capabilities is that firms can strike the organizational fit between in­
the supply chain context, many firms’ carbon emissions may stem from
formation processing requirements and information processing capa­
their suppliers (Villena and Dhanorkar, 2020). Hence, information on
bilities (Zhu, et al., 2018). In the supply chain context, firms can acquire
suppliers’ carbon emissions is beneficial to assess the carbon impact of
data-driven capabilities to collect, retain, and connect data from supply
the supply chain and direct initiatives for carbon emissions reduction
chain partners located both upstream and downstream (Kache and
(Luo, et al., 2018; Hahn et al., 2015). We follow several studies to
Seuring, 2017). These capabilities are widely proven to bring higher
establish a link between supply chain carbon transparency and two
visibility throughout the supply chain, enabling firms to be more
levels of performance: “straightforward” economic performance and
transparent in numerous supply chain sustainability issues such as GHG
“less self-evident” carbon performance of supply chains (Lintukangas,
emissions and climate change (Roβmann, et al., 2018; Fernando et al.,
et al., 2022; Shui et al., 2022; Sheng et al., 2023). Supply chain carbon
2018). Accordingly, as suggested by OIPT, alongside a large part of the
performance refers to the efficiency of efforts to reduce carbon emissions
existing body of knowledge, there are reasons to believe that data ana­
associated with the organization and supply chain processes. This en­
lytics capabilities could enhance carbon transparency by enabling firms
compasses the reduction of carbon emissions generated throughout
to proactively determine and alleviate carbon-related issues throughout
procurement, production, transportation, distribution, and other supply
their partners. Thus, we assume that.
chain and organization-related processes. On the other hand, supply
H1. Data analytics capabilities lead to higher supply chain carbon chain economic performance pertains to the efficiency of financial
transparency outcomes associated with the organization and supply chain activities.
On the other hand, several studies assert that business trans­ This includes procurement, production, distribution, and other supply
formation will likely lead to higher transparency at firm or supply chain chain and organization-related activities.
levels. For instance, the studies of Santharm and Ramanathan (2022) Data-driven digital transformation provides an advanced perspective
and Barrane et al. (2021) found a strong link between digitally enabled to effectively generate and use transparent carbon information
transformation and visibility and transparency throughout the supply throughout the supply chain. This is naturally a source of both economic
chain. Additionally, Akter et al. (2022) found that digital-driven inno­ and carbon performance. Indeed, despite the varying opinions among
vation in organizational transformation leads to supply chain trans­ researchers about the extent to which carbon transparency is relevant in
parency. Indeed, business transformation is an antecedent of several terms of economic and carbon performance, several studies provide
supply chain attributes positively related to greater supply chain reasons to postulate that carbon transparency, stemming from data-

4
A. Belhadi et al. International Journal of Production Economics 270 (2024) 109178

driven digital transformation, is a source of both economic and carbon In our qualitative research phase, we selected 13 focal firms repre­
performance. For instance, Liesen et al. (2017) posit that investors can senting diverse sectors within the supply chain industry. These sectors
attain risk-adjusted returns from firms that disclose their GHG emissions included but were not limited to, automotive, textiles, energy, and food
through analytical approaches, as opposed to those that do not. The production, encompassing a broad spectrum of supply chain dynamics.
authors contend that investment returns can be maximized by investing The firms were located across three distinct geographic contexts:
in firms that disclose a comprehensive level of GHG emissions, encom­ Europe, Africa, and Asia. This selection was intentional to ensure a
passing Scope 1, 2, and 3 emissions through a transformational comprehensive understanding of data-driven digital transformation
approach, in contrast to those that only disclose incomplete information. across varied supply chain environments. Firm size was categorized
Similar conclusions can be drawn from studies concerning carbon based on the annual revenue. Small firms had annual revenues of up to
neutrality, such as Qian and Schaltegger (2017), Hassan and Romilly $50 million, medium firms from $50 million to $500 million, and large
(2018), and Kim and Lyon (2011). Hence, there are enough reasons to firms over $500 million. Supply chain tiers were defined based on the
put forward. firm’s overall supply chain network position. Tier-1 firms are direct
suppliers to the final manufacturer, Tier-2 firms supply to Tier-1 firms,
H4. Supply chain carbon transparency mediates the relationship be­
and so on. The firms are selected as they have accomplished the data-
tween data-driven digital transformation and supply chain carbon
driven digital transformation according to our conceptualization. To
neutrality.
gain practical experience and better understand the phenomenon stud­
H5. Supply chain carbon transparency mediates the relationship be­ ied, we conducted semi-structured interviews with individuals in
tween data-driven digital transformation and supply chain economic various roles (CEOs, supply chain managers, and sustainability execu­
performance. tives). We observed the data-driven digital transformation in use at
different levels of the supply chain. Upon completing the interviews, we
3.3. The moderating effect of supply chain carbon uncertainties conducted a rigorous data analysis process. The transcribed interview
data underwent thematic analysis, a systematic approach that involved
In dynamic business environments, firms face uncertainties that coding segments of text to identify recurring patterns, concepts, and
require them to respond through investment and implementation of themes. This coding process involved carefully examining textual data,
information-acquiring and processing capabilities (Enrique, et al., 2022; where segments of interviews were systematically tagged and labeled to
Frank et al., 2022). Supply chain carbon uncertainties refer to the un­ capture specific ideas, concepts, or sentiments. The coded segments
certainties associated with estimating the carbon emissions and carbon were then organized into broader categories based on shared themes and
footprints of products and services across their entire supply chain commonalities. Through this iterative process, we identified recurring
(Mohammed, et al., 2017; Haddadsisakht and Ryan, 2018). Based on the patterns that elucidated the nuances of data-driven digital trans­
OIPT framework, our analysis suggests that companies facing supply formation, supply chain carbon neutrality, and economic performance
chain carbon uncertainties require increased information-processing within each firm and across different sectors and regions. The interviews
capabilities to make informed decisions. Data-driven digital trans­ provided qualitative insights to support our quantitative findings and
formation is especially relevant in such turbulent environments, as it aided in survey creation while minimizing bias. We were able to confirm
reinforces the importance of data analytics capabilities in a business the presence and relevance of certain concepts in the different countries
transformation perspective for achieving organizational goals (Chen and and supply chain sectors, identify primary data-driven digital trans­
Tian, 2022). The primary sources of supply chain uncertainty are up­ formation aspects, and determine appropriate variables to measure.
stream uncertainties caused by the variety of suppliers’ geographic lo­ Participants indicated that carbon neutrality at the supply chain level
cations and regulations (Mohammed, et al., 2017; Haddadsisakht and was their performance’s primary objective and indicator.
Ryan, 2018), and the opportunistic behavior of partners to hide or The quantitative phase involves surveying various supply chain
greenwash some pollutant activities (Kim and Lyon, 2011). Carbon sectors and seeking input from experts in the supply chain, operations,
uncertainties may constitute a boundary for the relationship between sustainability, and environmental management. The survey targeted
data-driven digital transformation and supply chain carbon trans­ high-level executives, directors, and managers in large manufacturing
parency. Indeed, in such uncertainties, the supply chain may be required firms located in Europe, Africa, and Asia. In selecting the countries for
to reinforce the data-driven digital transformation throughout the sup­ our study, we focused on Europe, Africa, and Asia due to their diverse
ply chain to achieve carbon transparency. Based on this, we propose the and representative nature in global supply chain dynamics. Europe was
following hypothesis. chosen for its advanced digital transformation initiatives and stringent
carbon footprint reduction policies. Africa was included to represent
H6.a. Supply chain carbon uncertainties negatively moderate the
emerging economies with growing digital transformation trends and
relationship between data-driven digital transformation and supply
increasing awareness of carbon footprint management. Asia was
chain carbon transparency.
selected for its significant role in global supply chains and its varied
Following similar logic, in the presence of carbon uncertainties,
approach to digital transformation and carbon reduction efforts. This
Carbon information should be spread over a significant disparity of
diversity provided a comprehensive perspective on the implementation
supply chain nodes with higher accuracy and clarity (Lintukangas, et al.,
and impact of data-driven digital transformation across different eco­
2022). Thus, firms are expected to attain higher levels of carbon trans­
nomic and environmental contexts. We targeted companies based on
parency to leverage carbon information to achieve both carbon
two criteria: i.e., a high level of data-driven digital transformation
neutrality and economic performance. Hence, we assume that.
adoption and a high level of activity in carbon footprint reduction across
H6.b. Supply chain carbon uncertainties negatively moderate the the supply chain. To ensure the validity of the survey, we first shared a
relationship between data-driven digital transformation and both car­ preliminary version with executives in the relevant fields and received
bon neutrality and the economic performance of the supply chain. feedback from 27 individuals. We then emailed the survey questionnaire
to our target population through the SurveyMonkey platform thrice
4. Research Methodology between January and May 2022. We targeted 1098 enterprises in the
three countries. These companies were identified through a business
4.1. Data collection association representing industrial supply chain activities. Finally, we
received 437 valid responses. The response rate is 39.8%. The main
Our research used a mixed-method approach that integrated quan­ characteristics and profiles of the sample enterprises are depicted in
titative and qualitative data to better comprehend the research context. Table 1.

5
A. Belhadi et al. International Journal of Production Economics 270 (2024) 109178

Table 1 reduction dummy variable based on the responses, which was equal to 1
Characteristics and profiles of the sample firms. if the manager reported answers (a) or (b) and 0 if they indicated an­
Firms # % Respondent profile # % swers (c), (d), or (e).
characteristics

Supply chain sector Department 4.2.2. Independent variables

Data-driven digital transformation. To operationalize data-driven
Automotive 87 19,91% General direction 39 8,92%
Airline 65 14,87% Supply chain 109 24,94%
digital transformation constructs, we first followed Srinivasan and
Textile 52 11,90% Operations 57 13,04% Swink (2018) and Akter et al. (2021) in operationalizing data analytics
Agrifood 49 11,21% Environment 105 24,03% capabilities as a firm’s data-driven analytics tools and techniques,
Energy 47 10,76% Sustainability 127 29,06% alongside the ability to leverage their output to inform the
Electric equipment 44 10,07% Experience
decision-making in the business transformation process. The measure­
Chemical 43 9,84% Less than five years 37 8,47%
Transportation 39 8,92% Between five and 15 103 23,57% ment is based on several factors, including their adeptness in utilizing
years statistical and data visualization techniques and their ability to effec­
Other 11 2,52% More than 15 years 297 67,96% tively deploy decision-making dashboards. In addition, we evaluated the
Firm’s size firm’s proficiency in gathering information from a wide range of sources
Small 22 5,03%
Medium 88 20,14%
and utilizing root cause analysis to dissect the data and identify un­
Large 327 74,83% derlying trends and issues. Second, we operationalized business trans­
Supply chain Tier formation based on the studies by Sheng et al. (2023) and Belhadi et al.
Focal firm 179 40,96% (2021). We used factors illustrating the depth of change made to orga­
Tier-1 135 30,89%
nizations and the firm’s business models.
Tier-2 84 19,22%
Tier-3 39 8,92% Supply chain carbon transparency. To assess supply chain carbon
Country transparency, we follow studies such as Shui et al. (2022), Schnacken­
Europe 167 38,22% berg et al. (2021), and Villena and Dhanorkar (2020) in using three
Asia 139 31,81% attributes: supply chain disclosure, information accuracy, and infor­
Africa 131 29,98%
mation clarity. First, we focus on information clarity as it is the key to
the transparency of carbon information. The process used to develop this
4.2. Measures variable is inspired by past studies by Villena and Dhanorkar (2020) and
Jira and Toffel (2013), which involved using the CDP’s questionnaire
Validated measurements from previous studies were utilized in this and 19 key questions answered by the firms. These questions are related
research. To establish the scale items, we used two versions, i.e., one in to identifying risks and opportunities linked to climate change, GHG
English in the Asian context and one in French in both French and emissions levels, and governance of climate change issues. Information
Moroccan contexts. Accordingly, the English version was translated into clarity was calculated by summing the number of questions answered by
French, and back-translation was employed to ensure cultural equiva­ the firm out of the 19, with a score of 10, for example, indicating that the
lence, as suggested by Sheng et al. (2023). Then, the measurement scale firm responded to 10 out of the 19 questions. Second, information ac­
has undergone a thorough review by both researchers and practitioners. curacy was assessed based on the percentage of uncertainties received in
A pre-test was conducted with 15 participants, and questionnaire carbon information at the supply chain level. An ordinal measure of
modifications were made based on their feedback. A seven-point Likert accuracy ranging from 0 to 10 was created using these percentages,
scale was used to measure the items. where lower uncertainty levels indicate higher accuracy. Third, supply
chain disclosure is critical to measuring carbon transparency because
4.2.1. Dependent variables firms can choose to disclose carbon information with all supply chain
In assessing the dependent variables of supply chain carbon perfor­ partners or just to the requesting partners (Villena and Dhanorkar,
mance and supply chain economic performance, our approach was 2020). This constitutes the primary measure of supply chain disclosure.
guided by established literature. Specifically, we developed a construct Supply chain carbon uncertainties: We followed studies such as
for supply chain carbon performance based on synthesizing existing Haddad-Sisakht and Ryan (2018) and Villena and Dhanorkar (2020) in
measures and definitions found in relevant studies. We adopted a measuring three dimensions of carbon-related uncertainties within the
multifaceted approach, integrating aspects such as energy usage, carbon supply chain. First, we included items to measure the level of un­
emissions, and utilization of carbon-intensive materials, as indicated by certainties in regulations and carbon thresholds. This is mainly related
Böttcher and Müller (2015) and Sheng et al. (2023). These measures to the country and region of the firm and the supply chain operations
were chosen for their relevance and consistency with the broader (Haddad-Sisakht and Ryan, 2018). Second, we measure the un­
conceptualization of supply chain carbon performance in current certainties caused by the fluctuations in demand and supply. This is
research, which views it as the effectiveness of supply chain operations mainly associated with unexpected changes in demand for products or
in minimizing carbon emissions. This includes (1) energy usage, (2) raw materials, which can result in shifts in supply chain activities. Third,
carbon emissions, and (3) carbon-intensive materials usage. Second, we include items to measure the inaccurate carbon accounting
supply chain economic performance was assessed using four measures throughout the supply chain.
following the study by Kim (2014), which are (1) supply chain costs as a
percentage of revenue, (2) Return on investment (ROI), (3) Inventory 4.2.3. Control variables
turnover, and (4) Perfect order rate. All the selected measures were We categorized control variables into four groups to mitigate biases
assessed in the percentage of variation (increase or decrease). To pre­ from omitted variables. For firm characteristics, we focused on size and
vent answers influenced by social desirability and bias towards a posi­ supply chain tier. Firm size was dichotomized (large firms as ’1′, small/
tive response, the dependent variable was operationalized by asking the medium as ’0′) to highlight differences in digital transformation ap­
respondents a multiple-choice question about the range of product unit proaches between larger and smaller firms. This dichotomy facilitated
costs in the preceding three years. The five possible answers were: (a) an analysis that sharply contrasts the impacts on and strategies of firms
less than a 10% reduction, (b) between a 10% and 3% reduction, (c) a of different sizes. Similarly, supply chain tiers were simplified into two
reduction or increase within 3%, (d) between a 3% and 10% increase, categories (focal firm and sub-tiers) to underline the varying roles and
and (e) more than a 10% increase. We computed an unbiased cost impacts at different supply chain levels. This allowed a more effective
comparison of digital transformation experiences across the supply

6
A. Belhadi et al. International Journal of Production Economics 270 (2024) 109178

chain. Second, we controlled for supply chain characteristics using one we followed Hayes and Rockwood (2017) in using the PROCESS macro
variable: complexity, which is associated with the degree of uncertainty, for the mediation effect and calculated indirect effects to assess media­
variability, and interdependence of the different components in the tion effects. As recommended by Hayes and Rockwood (2017), we used
supply chain. We used three dummies for three levels of supply chain a bootstrapping procedure with 5000 samples to test conditional indi­
complexities (simple, moderate, and complex). Third, we included two rect effects.
variables to control for country characteristics, i.e., institutional back­ Before conducting the analyses, we checked for assumptions such as
ground and technology development. To do that, we used two dummies linearity, homoscedasticity, normality, multicollinearity, and power
(where "1″ indicates institutional solid background or technology design (Cohen, 1992). We analyzed collinearity using partial re­
development and "0″ denotes medium to weak institutional background gressions, checked homoscedasticity visually, and assessed normality by
or technology development). The control procedure ensures the reli­ examining skewness and kurtosis parameters. We also checked the
ability and validity of our research findings, thereby upholding the variance inflation factor (VIF) for multicollinearity. We found the values
highest standards of professionalism in our study. of VIF to be less than 5 for all variables, which is acceptable following
the threshold of Hair et al. (2009). Finally, we used G-power analysis
4.3. Bias countermeasures following the procedure of Cohen (1992) to verify the feasibility of using
an OLS approach with our sample size (n = 437). Our sample size was
Using a firm-level survey where data on independent and dependent larger than the minimum necessary to achieve a statistically significant
variables were collected from a single respondent within each firm may power level.
generate a common method bias where data on independent and
dependent variables were collected from a single respondent within 5. Results
each firm (Doty and Glick, 1998). We employed procedural and meth­
odological approaches to minimize the effect of this bias. First, our re­ 5.1. Descriptive statistics
spondents were experienced and held mid/senior management
positions, which ensured that they possessed the required knowledge, Fig. 2 compares the adoption rates of data-driven digital trans­
thus mitigating some of the issues related to single-source bias (Chin, formation between the three countries (Europe, Africa, and Asia). The
et al., 2012). Additionally, we used separate sections for dependent and findings indicate that more than 50% of interviewed companies in the
independent variables in the survey design and ensured anonymity to three countries have implemented at least one data-driven digital
minimize social desirability bias. We also used indirect questions to transformation dimension: data analytics capabilities and business
mitigate social desirability bias (Malhotra, et al., 2006). Second, to transformation. However, the adoption rates of data analytics capabil­
further assess and minimize common method bias, we employed the ities and business transformation are higher in Europe and lower in
widely-used Harman’s single-factor test and conducted confirmatory Africa.
factor analysis by running a separate model incorporating a single latent
common method factor (CMF) (McFarlin and Sweeney, 1992; Sanchez
and Brock, 1996). The single latent CMF model yielded a χ2 = 1668.7 5.2. Hierarchical OLS regression analysis
with 191 degrees of freedom (compared with the χ2 = 285.6 and d.f. =
84 for the initial model). The results indicate that the fit of the single Our findings reveal a hierarchical structure comprising four inde­
latent CMF model was notably worse than the initial model, implying pendent models, each distinct model. We also adopt a unique structure
that common method bias is not a significant concern in this study. We for our mediation analysis, which follows Hayes’s (2017) method of
also used the Widaman (1985) approach to compare models with and computing indirect effects as a post-hoc analysis. Table 2 illustrates our
without CMF, which showed that the method factor accounted for only preliminary results from the regression analysis using the OLS proced­
1.62% of the total variance. Finally, we separated early and late re­ ure. Accordingly, we found that all the models at the final stages (i.e., 2,
sponses based on timestamps to address non-response bias concerns. In 3, and 4) were significant at 0.01.
addition, we conducted group comparison tests, which showed little The findings of Model 2 indicated that data analytics capabilities do
significant difference between the two groups in means and variations not significantly impact supply chain carbon transparency (β = 0.165, p
(p < 0.05). > 0.1), which does not support hypothesis H1. On the other hand,
business transformation positively influenced supply chain carbon
4.4. Data analysis transparency (β = 0.256, p < 0.01), supporting hypothesis H2. Inter­
estingly, this impact is amplified when combining business trans­
We conducted hierarchical Ordinary Least Squares (OLS) regression formation and data analytics capabilities in data-driven digital
analyses on our data to test our hypotheses. Hierarchical OLS regression
enabled us to assess the hierarchical relationships between variables,
which was suitable for addressing specific hypotheses related to the
influence of different factors step-wise. First, we standardized all inde­
pendent variables using mean-centering Z-scores before conducting the
analyses. Afterward, we ran three models, each of which began with a
control variables model in the first stage, followed by varying combi­
nations of data-driven digital transformation constructs, supply chain
carbon transparency constructs, and supply chain carbon uncertainty
constructs. The hierarchy of sets varies depending on each model. For
example, we included data analytics capabilities in the second stage and
business transformation in the third stage for supply chain carbon
transparency. In contrast, their combination was included in the fourth
stage. The fifth stage consists of the moderating effect of supply chain
carbon uncertainties. We included supply chain carbon transparency in
the second stage for supply chain carbon performance and supply chain
economic performance. In contrast, the moderating effect of supply Fig. 2. The adoption rates of data-driven digital transformation between the
chain carbon uncertainties was included in the third stage. Afterward, three countries.

7
A. Belhadi et al. International Journal of Production Economics 270 (2024) 109178

Fig. 3. Revised framework based on OIPT.

Table 2
Regression of hierarchical OLS regression analysis. ** <0.01, * <0.10.
Predictors Hypo. Supply chain carbon performance Supply chain economic
performance

Model Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Model 9 Model 10
1

F-value 1.523* 8.647** 13.736** 10.341** 8.369** 17.855** 6.946** 6.589** 1.769* 18.43**
R2 0.138 0.092 0.213 0.243 0.156 0.078 0.223 0.131 0.198 0.228
Adjusted R2 0.114 0.073 0.183 0.207 0.128 0.051 0.196 0.113 0.169 0.201
Direct relationships
Data analytics capabilities H1 0.165 0.148 0.151
Business transformation H2 0.256** 0.295** 0.307**
Data-driven digital transformation H3 0.418** 0.408**
Supply chain carbon transparency H4, 0.528** 0.497** 0.391** 0.447**
H5
Moderators
Supply chain carbon uncertainties x Data- H6.a − 0.117**
driven digital transformation
Supply chain carbon uncertainties x H6.b − 0.215** − 0.316**
Supply chain carbon transparency
Controls
Firm size 0.011 0.065 0.168 − 0.034* 0.167* 0.643 − 0.148* − 0.006 0.001 − 0.119
Firm tier 0.227 0.115* 0.005 − 0.016* 0.211* 0.518 − 0.098* − 0.002 − 0.102 − 0.113
Supply chain complexity 0.215 0.176* − 0.116* − 0.051* 0.216* 0.409 − 0.102* − 0.006 − 0.009 0.528
Institutional background 0.175 0.182* 0.007 0.000 0.184* 0.669 0.118* − 0.012 − 0.115 − 0.228
Technology development 0.184 0.119* 0.106 − 0.145* 0.173* 0.571 − 0.109* − 0.042 0.428* 0.288

transformation (β = 0.418, p < 0.01). This provides support for hy­ Regarding hypotheses H6, the findings of Models 7 and 10 provide
pothesis H3. statistical evidence of the negative moderating effect of supply chain
Furthermore, the results of Models 6 and 9 demonstrate a positive carbon uncertainties on both supply chain carbon performance (β =
impact of supply chain carbon transparency on both supply chain carbon − 0.215, p < 0.01) and supply chain economic performance (β = − 0.316,
performance (β = 0.528, p < 0.01) and supply chain economic perfor­ p < 0.01). Therefore, H6.a, and H6.b are supported.
mance (β = 0.391, p < 0.01).
To evaluate the mediating effect of supply chain carbon transparency
proposed in hypotheses H4 and H5, we employed the bootstrapping 5.3. Robustness checks
approach proposed by Hayes (2017). Table 3 depicts the mediation
analysis results. Our findings indicate that the constructs of supply chain To ensure the stability and consistency of our results, we conducted a
carbon transparency (i.e., information disclosure, information accuracy, series of robustness checks. We employed three distinct approaches to
and information clarity) have a significant and direct association with explore how the results of our regression analyses might vary. First, we
both supply chain carbon performance and supply chain economic introduced a new construct called "data analytics resources," which we
performance. This provides support to hypotheses H4 and H5. hypothesized to be an antecedent to enable data-driven digital trans­
formation. Using the confirmatory factor analysis approach, we

8
A. Belhadi et al. International Journal of Production Economics 270 (2024) 109178

Table 3 employed independently or in isolation —without being integrated into

The bootstrapping results of the mediating effect of supply chain carbon broader organizational strategies — these capabilities do not signifi­
transparency. cantly influence supply chain carbon transparency (Awan, et al., 2021;
Path Direct Indirect 95% confidence Total Enrique et al., 2022). However, their impact becomes markedly signif­
effect effect interval effect icant when these data analytics capabilities are incorporated into a more
Lower Upper comprehensive business transformation strategy, embracing organiza­
tional, cultural, and operational changes (Belhadi, et al., 2021). This
Data-driven digital 0.312** 0.129 0.002 0.512 0.442**
transformation → synergy between data analytics and holistic business transformation
Information disclosure profoundly influences supply chain carbon transparency. It allows for
→ Supply chain carbon effectively utilizing data to inform and drive sustainable practices
performance throughout the supply chain (Sheng, et al., 2023). This empirical reve­
Data-driven digital 0.259** 0.097 0.12 0.548 0.355
transformation →
lation underscores the symbiotic relationship between data analytics
Information accuracy capabilities and the comprehensive restructuring encapsulated within
→ Supply chain carbon business transformation—a hallmark of data-driven digital
performance transformation.
Data-driven digital 0.618** 0.175 0.544 0.811 0.793**
In addressing our second research question (RQ2), we explored how
transformation →
Information clarity → supply chain carbon transparency facilitates a balance between eco­
Supply chain carbon nomic performance and carbon neutrality, particularly under the
performance complexity of carbon uncertainties. Our analysis reveals a significant
Data-driven digital 0.217** 0.108 − 0.006 0.619 0.325** impact of supply chain carbon transparency on both carbon manage­
transformation →
Information disclosure
ment effectiveness and economic outcomes in the supply chain. This
→ Supply chain impact is evident through several key aspects of carbon transparency,
economic performance such as the thoroughness of information disclosure, accuracy, and
Data-driven digital 0.196 0.073 − 0.138 0.481 0.269** clarity (Villena and Dhanorkar, 2020). These aspects link strongly with
transformation →
the supply chain’s ability to manage carbon emissions and achieve
Information accuracy
→ Supply chain financial success. The role of supply chain carbon transparency emerges
economic performance as a crucial factor, effectively bridging the goals of economic growth and
Data-driven digital 0.284 0.115 0.008 0.519 0.399** carbon emission reduction. Additionally, our study highlights the sub­
transformation → stantial influence of carbon-related uncertainties in the supply chain.
Information clarity →
Supply chain
These uncertainties significantly shape how supply chains manage car­
economic performance bon emissions and perform economically, affecting the pursuit of carbon
neutrality and economic balance. Our findings illuminate the nuanced
role of carbon uncertainties in the supply chain, demonstrating their
measured “data analytics resources” with a five-item scale construct that effect on the interplay between carbon management and economic
included the necessary information, human resources, technological performance. This insight is critical in understanding the dynamics of
resources, financial resources, and organizational culture necessary to achieving carbon neutrality while maintaining economic viability in
develop data-driven digital transformation in the supply chain. Our supply chains. The findings of our study and the connection to the OIPT
assumption was confirmed when we added this construct to the models are illustrated in Fig. 3. The figure shows that the data-driven digital
where supply chain carbon transparency, economic performance, and transformation strategy is positioned at the top of the structure; indeed,
supply chain carbon performance were the dependent variables, as "data combining data analytics capabilities and business transformation
analytics resources" were not statistically significant in all models. strategy can provide a comprehensive organizational view to facilitating
Second, we analyzed the individual effect of each construct in our the implementation of data-driven digital transformation and support
models and found consistency with our main findings presented in increased carbon transparency and performance (Belhadi, et al., 2021;
Table 2. Finally, we utilized a financial performance construct as a Sheng et al., 2023). From the OIPT standpoint, this implies establishing
competing model to confirm that our primary model is not overfitted strategic alignment for an organizational information-processing fit, as
and that we have consistency in our analysis. As expected, the depicted in Fig. 3. In other words, companies must develop a data-driven
competing model was not supported, while our direct model showed digital transformation strategy that supports all the supply chain carbon
robust results. These procedures suggest that our models are consistent transparency dimensions (i.e., information disclosure, accuracy, and
and not overfitted and provide additional support for the validity and clarity) to create an information-processing capability that meets their
reliability of our findings. Overall, the robustness checks demonstrate information processing requirements (Villena and Dhanorkar, 2020).
the stability and consistency of our results, providing additional confi­ Our findings demonstrate that this is a fundamental requirement for
dence in the validity and reliability of our findings. achieving higher carbon transparency and carbon performance in the
supply chain amid carbon uncertain environments.
6. Discussion Interestingly, data analytics capabilities are unrelated to higher
supply chain carbon transparency. However, carbon neutrality is a
In this empirical study, we examine the impact of data-driven digital particular topic that needs cultural and organizational transformation to
transformation on economic and carbon performance through the OIPT leverage data analytics capabilities (Belhadi, et al., 2021). Indeed, these
lens while considering the mediating effect of supply chain carbon capabilities positively moderate the effect of a comprehensive business
transparency alongside the moderating effect of supply chain carbon transformation on transparency. Therefore, we expect combining data
uncertainties. The primary thrust of our first research question (RQ1) analytics capabilities and business transformation in the data-driven
revolves around delineating the intricate pathways through which data- digital transformation to provide more carbon transparency in the
driven digital transformation influences supply chain transparency. To supply chain. From an OIPT perspective, the connection between
address the query concerning the precise mechanisms underpinning this data-driven digital transformation and supply chain carbon trans­
influence, let us delve into the pragmatic facets of implementation. Our parency entails the organizational fit between information processing
analysis reveals a nuanced understanding of the role of data analytics capabilities and information processing requirements (Sheng, et al.,
capabilities in enhancing supply chain carbon transparency. When 2023). This fit can overcome the supply chain carbon uncertainties

9
A. Belhadi et al. International Journal of Production Economics 270 (2024) 109178

which constitute information processing boundaries to enhance supply possessing advanced data analytics tools is insufficient; their effective
chain performance at carbon and economic levels (Keller, et al., 2021). integration into organizational operations is critical. Managers should
The conclusions of our study transcend the mere recognition of the prioritize initiatives enhancing carbon transparency, including estab­
correlation between data-driven digital transformation and supply chain lishing robust data collection and reporting systems. This prioritization
carbon transparency as we explore the intricate relationship between is vital in light of uncertainties surrounding carbon neutrality in supply
data-driven digital transformation, economic performance, and carbon chains. By adopting such practices, managers can significantly improve
neutrality. In today’s supply chain landscape, firms face the daunting their ability to manage carbon emissions and meet carbon reduction
challenge of balancing the pursuit of economic performance with the targets. Additionally, recognizing the role of carbon transparency in
imperative of reducing carbon emissions (Villena and Dhanorkar, 2020). balancing economic performance with carbon reduction goals is crucial.
Our research reveals that the adoption of data-driven digital trans­ Managers should operationalize this by embedding sustainability met­
formation can serve as a means of resolving this dual pressure, with rics into performance assessments, prioritizing investments in carbon-
supply chain carbon transparency acting as an intermediary that effec­ efficient technologies, and fostering a culture of sustainability within
tively facilitates the improvement of economic and carbon performance their organizations. These actions are crucial for supply chain managers
in the supply chain (Sheng, et al., 2023; Belhadi et al., 2021). By to navigate the challenges of achieving carbon neutrality while main­
leveraging the potential of data analytics capabilities and business taining economic efficiency, as revealed by the study’s empirical find­
transformation, firms and supply chains can significantly enhance car­ ings. Implementing these recommendations will enable organizations to
bon transparency, balancing environmental and economic performance leverage better data-driven digital transformation for enhanced carbon
goals. transparency and performance in supply chains.

6.1. Theoretical implications 7. Conclusion

This study makes substantial theoretical contributions, particularly In this study, we have investigated the impact of data-driven digital
enriching the Organizational Information Processing Theory (OIPT) in transformation on balancing economic and carbon performance in the
data-driven digital transformation and supply chain carbon highly uncertain supply chain context. Accordingly, we have explored
management. the role of supply chain carbon transparency as a mediating factor be­
First, by empirically demonstrating the interplay between data- tween data-driven digital transformation and the simultaneous
driven digital transformation and supply chain carbon transparency, improvement of economic and carbon performance. As a result, we
this research advances our understanding of the OIPT framework (Gal­ found that data analytics capabilities are only sufficient to enhance the
braith, 1974). Contrary to the primary theoretical focus of prior studies supply chain carbon uncertainties once included in the broader business
(e.g., Awan et al., 2021; Belhadi et al., 2021), our findings provide transformation, enabling the deployment of the necessary skills, infra­
concrete evidence that integrating data analytics capabilities with structure, and processes. Moreover, supply chain carbon transparency is
broader business transformation is critical for enhancing carbon trans­ found to overcome the negative effect of carbon uncertainties and sup­
parency. This novel insight extends OIPT by showcasing how informa­ port data-driven digital transformation in simultaneously improving
tion processing capabilities, a core element of the theory, can be supply chains’ economic and carbon performance.
strategically aligned with environmental sustainability goals. Second, Although our study contributes to the literature on data-driven dig­
the study offers fresh perspectives on the synergistic impact of ital transformation and supply chain carbon neutrality, it also has some
data-driven digital transformation on both economic and carbon per­ limitations that suggest promising avenues for future research. First, we
formance, bridging a gap noted in previous research (Chen and Tian, used cross-sectional data to examine the mediating role of supply chain
2022; Enrique et al., 2022; Papanagnou et al., 2022). Our findings carbon transparency, which precludes establishing causal relationships
contribute to the OIPT by illustrating how enhanced information pro­ and limits our ability to assess the long-term effects of carbon trans­
cessing capabilities can improve economic efficiency and carbon parency. Hence, future studies employing longitudinal research designs
reduction, offering a more holistic view of organizational performance. could provide a more comprehensive understanding of the relationships
Third, our research enriches the OIPT (Galbraith, 1974; Zhu et al., 2018) between digital data-driven transformation, carbon transparency, and
by exploring how data-driven digital transformation aids in navigating performance. Second, our study found that supply chain carbon un­
supply chain carbon uncertainties. We demonstrate that effective in­ certainties weaken the impact of data-driven digital transformation on
formation processing, central to OIPT, is critical in achieving supply supply chain carbon transparency due to the vital need for information
chain carbon transparency amidst these uncertainties. This extends the processing capabilities. Future research could explore additional key
application of OIPT to address the contemporary challenges of envi­ digital features that can mitigate the impact of carbon uncertainties
ronmental sustainability in supply chains, providing a nuanced under­ using different theoretical perspectives, such as the dynamic capabilities
standing of how information processing capabilities underpin successful perspective and resource-based view. Third, we used survey data
carbon management strategies. collected from firms from three countries, i.e., Africa, Europe, and Asia,
In sum, our study not only responds to calls for empirical examina­ which can limit the generalizability of our findings to other contexts.
tion within these domains (Belhadi, et al., 2021; Pappas, 2018) but also Therefore, future studies could replicate our study in diverse settings
significantly contributes to the OIPT by demonstrating its applicability and extend the scope of inquiry to include other types of firms, such as
and relevance in the context of digital transformation and carbon service providers and retailers. These avenues for future research could
management in supply chains. shed further light on the complex relationships between data-driven
digital transformation, supply chain carbon transparency, and
6.2. Managerial implications performance.

Based on our research findings, we provide the following recom­ CRediT authorship contribution statement
mendations for supply chain practitioners. The study highlights the
imperative for supply chain managers to strategically adopt data-driven Amine Belhadi: Conceptualization, Data curation. Mani Ven­
digital transformation strategically, focusing on integrating data ana­ katesh: Conceptualization, Methodology, Supervision, Writing – review
lytics capabilities within broader business strategies. This approach is & editing. Sachin Kamble: Data curation, Formal analysis, Investiga­
essential for improving supply chain carbon transparency, a critical tion, Resources, Validation, Writing – original draft. Mohammad Zay­
factor in achieving carbon neutrality. The findings suggest that merely nol Abedin: Data curation, Formal analysis, Project administration,

10
A. Belhadi et al. International Journal of Production Economics 270 (2024) 109178

Supervision, Writing – review & editing. Hassan, O.A., Romilly, P., 2018. Relations between corporate economic performance,
environmental disclosure and greenhouse gas emissions: new insights. Bus. Strat.
Environ. 27 (7), 893–909.
Data availability Hayes, A.F., 2017. Introduction to Mediation, Moderation, and Conditional Process Analysis:
A Regression-Based Approach. s.L. Guilford publications.
The authors do not have permission to share data. Hayes, A.F., Rockwood, N.J., 2017. Regression-based statistical mediation and
moderation analysis in clinical research: Observations, recommendations, and
implementation. Behav. Res. Ther. 98, 39–57.
Acknowledgement Jira, C.(., Toffel, M.W., 2013. Engaging supply chains in climate. Manuf. Serv. Oper.
Manag. 15 (4), 523–700.
Kache, F., Seuring, S., 2017. Challenges and opportunities of digital information at the
We acknowledge the University of Manitoba’s start up research grant intersection of Big Data Analytics and supply chain management. Int. J. Oper. Prod.
no 326917-322017-2000 for the support. Manag. 37 (1), 10–36.
Keller, J., Burkhardt, P., Lasch, R., 2021. Informal governance in the digital
transformation. Int. J. Oper. Prod. Manag. 41 (7), 1060–1084.
References Kim, D.-Y., 2014. Understanding supplier structural embeddedness: a social network
perspective. J. Oper. Manag. 32 (5), 219–231.
Aben, T.A., Valk, W.v. d., Roehrich, J.K., Selviaridis, K., 2021. Managing information Kim, E.-H., Lyon, T.P., 2011. Strategic environmental disclosure: evidence from the
asymmetry in public–private relationships undergoing a digital transformation: the DOE’s voluntary greenhouse gas registry. J. Environ. Econ. Manag. 61 (3), 311–326.
role of contractual and relational governance. Int. J. Oper. Prod. Manag. 41 (7), Liesen, A., Figge, F., Hoepner, A., Patten, D.M., 2017. Climate change and asset prices:
1145–1191. are corporate carbon disclosure and performance priced appropriately? J. Bus.
Akter, S., Bandara, R.J., Sajib, S., 2021. How to empower analytics capability to tackle Finance Account. 44 (1–2), 35–62.
emergency situations? Int. J. Oper. Prod. Manag. 41 (9), 1469–1494. Lintukangas, K., Arminen, H., Kähkönen, A., Karttunen, E., 2022. Determinants of supply
Akter, S., et al., 2022. Transforming business using digital innovations: the application of chain Engagement in carbon management. J. Bus. Ethics.
AI, blockchain, cloud and data analytics. Ann. Oper. Res. 308, 7–39. Luo, L., Tang, Q., Peng, J., 2018. The direct and moderating effects of power distance on
Akter, S., et al., 2016. How to improve firm performance using big data analytics carbon transparency: an international investigation of cultural value and corporate
capability and business strategy alignment? Int. J. Prod. Econ. 182, 113–131. social responsibility. Bus. Strat. Environ. 27 (8), 1546–1557.
Awan, U., et al., 2021. Big data analytics capability and decision-making: the role of Malhotra, N.K., Kim, S.S., Patil, A., 2006. Common method variance in IS research: a
data-driven insight on circular economy performance. Technol. Forecast. Soc. comparison of alternative approaches and a reanalysis of past research. Manag. Sci.
Change 168. 52 (12), 1865–1883.
Barrane, F.Z., et al., 2021. Building trust in multi-stakeholder collaborations for new Matthias, O., Fouweather, I., Gregory, I., Vernon, A., 2017. Making sense of Big Data –
product development in the digital transformation era. Benchmark Int. J. 28 (1), can it transform operations management? Int. J. Oper. Prod. Manag. 37 (1), 37–55.
205–228. McFarlin, D.B., Sweeney, P.D., 1992. Distributive and procedural justice as predictors of
Belhadi, A., Kamble, S., Gunasekaran, A., Mani, V., 2021. Analyzing the mediating role of satisfaction with personal and organizational outcomes. Acad. Manag. J. 35 (3),
organizational ambidexterity and digital business transformation on industry 4.0 626–637.
capabilities and sustainable supply chain performance. Supply Chain Manag.: Int. J. Mikalef, P., Boura, M., Lekakos, G., Krogstie, J., 2019. Big data analytics capabilities and
27 (6), 696–711. innovation: the mediating role of dynamic capabilities and moderating effect of the
Böttcher, C.F., Müller, M., 2015. Drivers, practices and outcomes of low-carbon environment. Br. J. Manag. 30 (2), 272–298.
operations: approaches of German automotive suppliers to cutting carbon emissions. Mohammed, F., Selim, S.Z., Hassan, A., Syed, M.N., 2017. Multi-period planning of
Bus. Strat. Environ. 24 (6), 477–498. closed-loop supply chain with carbon policies under uncertainty. Transport. Res.
Chen, H., Tian, Z., 2022. Environmental uncertainty, resource orchestration and digital Transport Environ. 51, 146–172.
transformation: a fuzzy-set QCA approach. J. Bus. Res. 139, 184–193. Papadopoulos, T., Balta, M.E., 2022. Climate Change and big data analytics: challenges
Chin, W.W., Thatcher, J.B., Wright, R.T., 2012. Assessing common method bias: and opportunities. Int. J. Inf. Manag. 63.
Problems with the ULMC technique. MIS Q. 36 (3), 1003–1019. Papanagnou, C., et al., 2022. Data-driven digital transformation for emergency
Cohen, J., 1992. Statistical power analysis. Curr. Dir. Psychol. Sci. 1 (3), 98–101. situations: the case of the UK retail sector. Int. J. Prod. Econ. 250.
Doty, D.H., Glick, W.H., 1998. Common methods bias: does common methods variance Pappas, I. O. P. M. M. N. G. J. K. a. G. L., 2018. Big data and business analytics
really bias results? Organ. Res. Methods 1 (4), 374–406. ecosystems: paving the way towards digital transformation and sustainable societies.
Enrique, D.V., et al., 2022. Being digital and flexible to navigate the storm: how digital Inf. Syst. E Bus. Manag. 16, 479–491.
transformation enhances supply chain flexibility in turbulent environments. Int. J. Qian, W., Schaltegger, S., 2017. Revisiting carbon disclosure and performance:
Prod. Econ. 250. Legitimacy and management views. Br. Account. Rev. 49 (4), 365–379.
Fernando, Y., Chidambaram, R.R., Wahyuni-Td, I.S., 2018. The impact of Big Data Roßmann, B., Canzaniello, A., Gracht, H.v. d., Hartmann, E., 2018. The future and social
analytics and data security practices on service supply chain performance. impact of big data analytics in supply chain management: results from a Delphi
Benchmark Int. J. 25 (9), 4009–4034. study. Technol. Forecast. Soc. Change 130, 135–149.
Filom, S., Amiri, A.M., Razavi, S., 2022. Applications of machine learning methods in Sanchez, J.I., Brock, P., 1996. Outcomes of perceived discrimination among Hispanic
port operations–A systematic literature review. Transport. Res. E Logist. Transport. employees: is diversity management a luxury or a necessity? Acad. Manag. J. 39 (3),
Rev. 161. 704–719.
Fosso Wamba, S., Queiroz, M.M., Guthrie, C., Braganza, A., 2022. Industry experiences of Santharm, B.A., Ramanathan, U., 2022. Supply chain transparency for sustainability – an
artificial intelligence (AI): benefits and challenges in operations and supply chain intervention-based research approach. Int. J. Oper. Prod. Manag. 42 (7), 995–1021.
management. Prod. Plann. Control 33 (16), 1493–1497. Schnackenberg, A.K., Tomlinson, E.C., Coen, C., 2021. The dimensional structure of
Frank, A.G., Mendes, G.H.d.S., Benitez, G.B., Ayala, N.F., 2022. Service customization in transparency: a construct validation of transparency as disclosure, clarity, and
turbulent environments: service business models and knowledge integration to accuracy in organizations. Journal of management 74 (10), 1628–1660.
create capability-based switching costs. Ind. Market. Manag. 100, 1–18. Shafiq, A., Ahmed, M.U., Mahmoodi, F., 2020. Impact of supply chain analytics and
Galbraith, J.R., 1974. Organization design: an information processing view. Interfaces 4 customer pressure for ethical conduct on socially responsible practices and
(3), 28–36. performance: an exploratory study. Int. J. Prod. Econ. 225.
Gong, C., Ribiere, V., 2021. Developing a unified definition of digital transformation. Sheng, H., Feng, T., Liu, L., 2023. The influence of digital transformation on low-carbon
Technovation 102. operations management practices and performance: does CEO ambivalence matter?
Govindarajan, V., Srivastava, A., Venkatesh, M., 2021. How Working with Competitors Int. J. Prod. Res.
made Jio a Telecom Giant: what other companies can learn from its success. Harv. Shui, X., Zhang, M., Smart, P., 2022. Climate Change Disclosure and the Promise of
Bus. Rev. 20 dec 2021.Accessed on 15/11/2023 This does not have volume or page Response-Ability and Transparency: A Synthesizing Framework and Future Research
as it is HBR article https://hbr.org/2021/12/how-working-with-competitors-made-ji Agenda. European Management Review.
o-a-telecom-giant. Simon, H.A., 1957. Models of Man; Social and Rational. s.L. Wiley.
Haddadsisakht, A., Ryan, S.M., 2018. Closed-loop supply chain network design with Srinivasan, R., Swink, M., 2018. An investigation of visibility and flexibility as
multiple transportation modes under stochastic demand and uncertain carbon tax. Complements to supply chain analytics: an organizational information processing
Int. J. Prod. Econ. 195, 118–131. theory perspective. Prod. Oper. Manag. 27 (10), 1849–1867.
Hahn, R., Reimsbach, D., Schiemann, F., 2015. Organizations, climate change, and Villena, V.H., Dhanorkar, S., 2020. How institutional pressures and managerial
transparency: Reviewing the literature on carbon disclosure. Organ. Environ. 28 (1), incentives elicit carbon transparency in global supply chains. J. Oper. Manag. 66 (6),
80–102. 697–734.
Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E., 2009. Multivariate Data Analysis, 7th Walmart Sustainability Hub, 2020. Project Gigaton [Online]. https://www.walmartsus
éd. Prentice Hall, Indian: s.n. tainabilityhub.com/climate/project-gigaton. (Accessed 2 April 2023).
Hanelt, A., Bohnsack, R., Marz, D., Marante, C.A., 2021. A systematic review of the Widaman, K.F., 1985. Hierarchically nested covariance structure models for multitrait-
literature on digital transformation: insights and Implications for strategy and multimethod data. Appl. Psychol. Meas. 9 (1), 1–26.
organizational change. J. Manag. Stud. 58 (5), 1159–1440. Zhu, S., et al., 2018. How supply chain analytics enables operational supply chain
transparency. Int. J. Phys. Distrib. Logist. Manag. 48 (1), 47–68.

11