Path Analysis of Supply Chain Management Reform Driven by Data

Empowerment

Ruijin Xu

School of Management, UCL, E14 5AA, London, U.K.

Keywords: Artificial Intelligence, Supply Chain Centralization, Green Supply Chain Management, Unilever, Hailan

Home.

Abstract: Amid converging digitalization and sustainability imperatives, this study advances a conceptual framework

integrating artificial intelligence (AI)-enabled analytics, supply chain centralization, and green supply chain

management (GSCM) as mutually reinforcing drivers of supply chain performance. The framework posits

that when synergistically combined with AI and environmental imperatives, centralized governance structures

generate superior efficiency-sustainability outcomes. Empirical validation was provided through a

comparative case analysis of Unilever and Hailan Home (HLA), illustrating divergent integration pathways.

Unilever’s globally centralized, data-driven, and sustainability-aligned supply chain strategy contrasts HLA’s

efficiency-focused model. Results demonstrate that triadic integration facilitates enhanced visibility,

operational agility, and environmental performance. The findings contribute to SCM theory by aligning

resource-based and dynamic capabilities perspectives with environmental strategy while offering actionable

managerial insights on orchestrating digital, structural, and green initiatives. The study concludes that

strategic alignment across technological, organizational, and ecological dimensions is foundational for

competitive resilience in data-driven supply ecosystems.

1 INTRODUCTION

Global supply chains today face dual pressures to

improve operational efficiency while meeting

sustainability targets amid rapid digitalization (De et

al., 2016). Firms must cut costs and boost speed to

stay competitive even as stakeholders demand greater

environmental and social responsibility throughout

the supply chain (De et al., 2016). These twin

imperatives have fueled interest in data-driven supply

chain management-leveraging big data analytics and

AI for decision-making—and green supply chain

management (GSCM) that integrates sustainability

into SCM practices. At the same time, companies are

reconsidering structural strategies; for example, many

are pursuing supply chain centralization to gain better

control and economies of scale in a volatile global

market. This study is motivated by the need to

understand how these three themes-AI-driven

decision-making, centralization, and GSCM-can be

integrated to achieve a "double win" of high

performance and sustainability.

https://orcid.org/0009-0006-6856-5261

This research addresses the following questions.

First, how can a centralized, AI-empowered supply

chain enhance operational efficiency and

environmental sustainability. Secondly, does

increasing supply chain centralization lower costs and

improve performance, and under what conditions.

Third, how do GSCM practices impact firm

performance-can companies "go green" and still

improve profitability. Finally, are AI-based supply

chain analytics (e.g., demand forecasting and

inventory management) superior to traditional

methods, and how do they influence outcomes like

service levels and waste reduction. By examining

these questions together, the study addresses a gap in

prior research – most studies have considered these

strategies in isolation, whereas their interaction may

unlock additional performance gains (Nguyen et al.,

2022).

The contributions are twofold. Theoretically, this

work develops an integrative framework linking

SCM strategy (centralization vs. decentralization),

technological innovation (AI and data analytics), and

270

Xu, R.

Path Analysis of Supply Chain Management Reform Driven by Data Empowerment.

DOI: 10.5220/0013842600004719

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 2nd International Conference on E-commerce and Modern Logistics (ICEML 2025), pages 270-276

ISBN: 978-989-758-775-7

sustainability (GSCM), extending supply chain

management theory to encompass their interplay.

Practically, it analyses two industry cases-Unilever

(fast-moving consumer goods) and HLA (apparel

retail)-to illustrate how organizations can implement

these strategies in tandem. The findings offer

actionable insights into managers redesigning supply

chains to be more data-driven and green while

maintaining cost-effectiveness. In sum, efficiency

and sustainability need not be a trade-off; with the

strategic integration of AI, structural centralization,

and green practices, firms can achieve synergistic

improvements in supply chain performance.

2 LITERATURE REVIEW

In contemporary supply chain scholarships, artificial

intelligence (AI), supply chain centralization, and

green supply chain management (GSCM) are

prominent, interrelated paradigms for enhancing

performance. AI-driven analytics has improved

demand forecasting, inventory management, and

responsiveness (Arunachalam et al., 2018; Dubey et

al., 2019). Mäenpää (2024) contends that AI has

become essential for harnessing big data and

embedding predictive intelligence into supply chain

decision-making. Concurrently, centralizing supply

chain control yields coordination benefits through

risk pooling and economies of scale, as formally

quantified by Eppen (1979) in classical inventory

theory and demonstrated in practice by Brito (2016).

However, excessive centralization can diminish local

agility; thus, scholars advocate balancing central

authority with decentralized flexibility (Qi et al.,

2017).

GSCM extends traditional SCM by integrating en

vironmental sustainability across all supply chain act

ivities (Srivastava, 2007). A robust body of literature

finds that GSCM practices often jointly enhance env

ironmental and operational performance, challenging

the notion of an efficiency–sustainability trade-off (

Zhu & Sarkis, 2004; Kalyar et al., 2019). For exampl

e, companies adopting eco-efficient processes have r

ealized cost savings and revenue gains from green pr

oducts (Feng et al., 2017). Researchers also note imp

lementation challenges, including higher upfront cos

ts and supplier coordination complexities (Mudgal et

al., 2010; Walker et al., 2008). Although some have

posited the “fallacy of profitable green supply chains

”-the idea that sustainability gains may initially come

at the expense of profitability (Esfahbodi et al., 2023

)-the prevailing view is that well-executed GSCM ult

imately bolsters long-term efficiency and resilience (

Green et al., 2012; Zhu et al., 2013) while advancing

sustainability goals. Moreover, these three paradigm

s are increasingly regarded as complementary. AI an

d centralized data systems can amplify GSCM outco

mes by optimizing resource use and enabling agile, s

ustainable decision-making, thereby aligning efficien

cy with environmental responsibility.

3 INTEGRATIVE THEORETICAL

SYNTHESISM

Synthesizing the discussion, a conceptual framework

interlinks AI-driven decision-making, supply chain

centralization, and green supply chain management

(GSCM) as core elements of a data-driven sustainable

supply chain strategy. While each component

independently influences performance, their

integration yields synergistic benefits and mitigates

trade-offs that arise in isolation.

A key interplay exists between AI capabilities and

a centralized supply chain structure. Centralization

entails reduced flexibility, but AI-driven analytics can

mitigate this drawback by enabling the centralized

system to sense and respond to local variations. In a

centralized model, large volumes of data from all

regions are funded into a single hub; AI is essential to

process this data in real time and provide decision-

makers with location-specific insights. An AI-

enabled control tower can detect regional demand

shifts and adjust plans, accordingly, imbuing a

centralized system with adaptability akin to

decentralization. Moreover, centralization provides a

unified IT infrastructure-a coherent data repository

for algorithms-and a clear mandate for innovation,

factors that expedite AI deployment. In resource-

based view (RBV) terms, combining a centralized

structure with AI capabilities creates a complex, hard-

to-imitate resource that confers a sustainable

competitive advantage (Barney, 1991). A positive

interaction is evident: centralization offers the scale

and data coherence on which AI thrives, while AI

reduces centralization’s risks by enabling data-driven

agility.

AI enables GSCM, equipping firms with tools to

monitor and optimize environmental performance.

Firms integrating AI into GSCM can use it for real-

time carbon footprint tracking and pinpointing

inefficiencies driving emissions. AI-driven

optimization models balance cost and ecological

objectives, finding sourcing, production, and logistics

solutions that might elude human planners. Qu and

Kim (2024) observe that current AI applications in

Path Analysis of Supply Chain Management Reform Driven by Data Empowerment

271

sustainable SCM mainly address environmental and

economic goals, while social dimensions lag (Qu &

Kim, 2024). This suggests AI is well-suited to pursue

cost efficiency and environmental impact reduction

jointly—the precise overlap of GSCM targets. In this

framework, AI augments GSCM by managing

complexity (e.g., life-cycle assessment data) and

enabling continuous tracking of green performance

metrics. Conversely, GSCM priorities (e.g., reducing

carbon or hazardous material use) spur novel AI

applications beyond traditional cost and service

metrics.

Centralization and GSCM are complementary: a

centralized structure facilitates uniform implementati

on of environmental standards across the organizatio

n. A central procurement function can enforce green

policies enterprise-wide, whereas decentralized units

might lag in compliance. Unilever’s centralized Eur

opean logistics network enabled a concerted push tha

t cut transport CO₂ emissions by 20% over five years

(De et al., 2016), aligning logistics operations with c

orporate sustainability targets. Centralization can, ho

wever, also concentrate on environmental impact (e.

g., one large warehouse may increase transport dista

nces). However, judicious network design (e.g., regi

onal hubs and optimized routing) can offset this by r

educing total miles travel, increasing load efficiency,

and lowering emissions per unit.

AI, centralization, and GSCM push the

performance frontier to new heights when

implemented in tandem. Centralization with AI

markedly improves operational metrics (speed, cost,

service); centralization with GSCM enhances

environmental performance without undermining

efficiency; and AI with GSCM ensures that

sustainability improvements are achieved cost-

effectively. Thus, at the intersection of all three, a

firm can attain strong economic performance

alongside strong environmental performance-a win–

win scenario.

This integrative framework aligns with emerging

perspectives in strategic supply chain management

and operations research that emphasize multifaceted

capabilities. For instance, the notion of a “digital

supply chain for sustainability” suggests leveraging

digitalization (AI, big data) for sustainability

objectives, not just operational goals (Büyüközkan &

Göçer, 2018). The framework extends this concept by

incorporating a centralization dimension. It also

aligns dynamic capabilities theory, which posits that

firms must continually adapt structures, technologies,

and objectives to sustain performance in changing

environments.

Integrating AI, supply chain centralization, and

green practices yield a holistic approach that drives

superior operational and environmental performance.

The following case analyses of Unilever and Hailan

Home (HLA) demonstrate how this integrated

framework operates in practice, providing empirical

insight into its benefits and trade-offs.

4 CASE ANALYSIS

4.1 Unilever: Data-Driven

Centralization and Green SCM on

a Global Scale

Unilever is among the world’s most significant fast-

moving consumer goods (FMCG) companies,

producing food, home care, and personal care

products sold in over 190 countries. By the mid-

2000s, Unilever’s European supply chain was highly

fragmented along national lines, resulting in

duplication and suboptimal asset utilization. Around

2008, Unilever launched Ultralogistik, a significant

initiative to centralize European logistics via a single

control tower and a network of regional distribution

hubs. In parallel, Unilever emerged as a corporate

sustainability leader, introducing its Sustainable

Living Plan in 2010 with ambitious environmental

targets (e.g., halving the environmental footprint of

its products). By the late 2010s, the company also

began substantial investments in digital supply chain

capabilities, including predictive analytics for

demand forecasting and optimization tools, to further

streamline operations.

4.1.1 Centralisation Strategy

Under Ultralogistik, Unilever consolidated transport

planning and warehouse management across multiple

European countries. Rather than each nation

managing logistics, a centralized control tower

coordinated shipments across markets. This

restructuring yielded classic efficiency gains: service

levels improved, and costs declined markedly. Over

five years, Unilever’s on-time in-full delivery

performance increased from ~97.5% to 98.8%,

accompanied by a 20% reduction in transportation

CO₂ emissions and roughly €91 million in annual

logistics cost savings (De et al., 2016). These

improvements were directly attributed to centralized

coordination. For example, cross-market truck

consolidation minimized empty runs, and risk

pooling at central warehouses optimized inventory

levels by eliminating inefficiencies hidden in siloed

ICEML 2025 - International Conference on E-commerce and Modern Logistics

272

national operations (De et al., 2016). Centralization

also enhanced visibility: a unified system allowed

Unilever to track end-to-end performance metrics and

identify bottlenecks that decentralized systems had

obscured (De et al., 2016). Furthermore, the company

was able to standardize best practices across the

network. Unilever reported benefits such as greater

purchasing leverage and a unified delivery process

leading to fewer errors and delays, confirming

theoretical expectations that centralization yields

economies of scale and consistency (Unilever, 2015b；

De et al., 2016).

4.1.2 Data-Driven and AI Initiatives

Having established a centralized data infrastructure,

Unilever aggressively incorporated advanced

analytics and AI into its supply chain operations. The

company developed sophisticated demand

forecasting systems utilizing machine learning to

factor in promotions, weather patterns, and social

media trends. In the ice cream category, where

demand is highly weather-dependent-AI-driven

models now adjust production and inventory based on

temperature forecasts, mitigating overstock during

cool periods and preventing stockouts during heat

waves. According to company reports, this dynamic,

weather-responsive planning helped “cut waste”

significantly by ensuring ice creams do not sit unsold

in unseasonable conditions. Unilever has also

deployed IoT sensors with AI analytics in over

100,000 retail freezers; these freezers transmit real-

time stock and performance data to a central platform,

where analysis identified opportunities to optimize

product placement, yielding a 30% increase in sales

in specific markets. This exemplifies how

centralizing data collection and applying AI-driven

insights can deliver substantial performance gains.

Beyond these initiatives, Unilever utilizes AI for

supply chain planning and optimization, using

algorithms to schedule production and distribution to

minimize cost while meeting service targets.

Internally, the company reports that AI-driven

decision systems have improved forecast accuracy by

15–20% and reduced overall inventory levels by a

similar margin without compromising customer

service (Unilever, 2022).

4.1.3 GSCM and Sustainability

Sustainability is deeply ingrained in Unilever’s

supply chain strategy. The company has a

comprehensive GSCM program that extends from

sustainable sourcing of raw materials (e.g., certified

palm oil, tea) to eco-efficient manufacturing (steady

reductions in factory energy, water, and waste) to

green logistics. By centralizing logistics, Unilever

implemented a low-carbon transport initiative,

optimizing truckloads and routes with AI tools and

substantially reducing transport emissions (De et al.,

2016). Additionally, Unilever piloted alternative

fuels (e.g., biofuels) and modal shifts from road to rail

as part of its centralized planning – efforts that

required cross-border coordination, which the central

model facilitated. On the distribution side, Unilever’s

central team pursued packaging reduction and

increased the recyclability of shipping materials. The

sustainability results of these efforts have been

notable. Alongside cost savings, Unilever cut over

15,000 tonnes of CO₂ annually from its European

logistics network by consolidating loads and adopting

greener transport modes (Brito, 2016). Globally, by

2020, Unilever had eliminated non-hazardous waste

to landfill in its manufacturing operations and was

sourcing a large share of its electricity from

renewables– accomplishments driven by policies

executed through centralized oversight. One

illustrative outcome is that Unilever’s “Sustainable

Living” brands (its product lines most fully aligned

with sustainability objectives) accounted for 75% of

the company’s growth in 2018 and grew 69% faster

than other brands. While this extends beyond supply

chain operations, coupling sustainability with core

business strategy can yield competitive advantages.

From a supply chain perspective, Unilever

successfully integrated GSCM into its centralized,

data-driven model – for example, centrally enforcing

stringent emissions standards for all logistics partners

and tracking compliance via data systems (Smart

Freight Centre, 2021).

4.1.4 Performance Outcomes

Unilever’s holistic strategy has led to quantitative and

qualitative performance improvements.

Quantitatively, as noted, service levels (OTIF ~98.8%)

and cost efficiency improved post-centralization (De

et al., 2016); inventory levels relative to sales

declined (inventory turnover increased), and supply

chain carbon emissions were substantially reduced.

Qualitatively, the supply chain became more agile

through data-driven scenario planning and rapid

responsiveness to market changes. Moreover,

Unilever’s reputation and brand equity benefited

from its demonstrated leadership in sustainable

supply chain practices, engendering goodwill among

customers and investors. Overall, Unilever’s case

indicates that concurrently deploying AI,

centralization, and GSCM can create mutually

Path Analysis of Supply Chain Management Reform Driven by Data Empowerment

273

reinforcing benefits, resulting in a resilient, efficient,

and environmentally responsible supply chain-albeit

one that requires significant investment and

organizational commitment to develop.

4.2 Hailan Home (HLA): Centralized

and Tech-Enabled Retail Supply

Chain with a Focus on Efficiency

4.2.1 Background

Hailan Home Co., Ltd. (HLA) is one of China’s

largest menswear retailers, operating a unique model

with thousands of stores (primarily franchised or

affiliated) nationwide. By the end of 2021, HLA had

over 7,300 stores nationwide (approximately 6,579

franchise or affiliate stores and the rest self-operated)

(Chen, 2022). China’s apparel retail market is

intensely competitive and has faced rising labor and

material costs and increasingly diverse consumer

preferences. HLA attributed its rapid growth and

success to relentless cost control and supply chain

efficiency in this environment. A core element of

HLA’s strategy is supplying chain centralization,

particularly creating a unified logistics and inventory

management system to support its extensive store

network.

4.2.2 Centralization Strategy

HLA established a single central distribution park in

Jiangsu Province to serve as the primary logistics hub

for all stores. Through this hub and a “direct

management” approach to franchise stores, HLA

maintains centralized ownership of inventory at the

headquarters level (a consignment model) and

directly manages store replenishment.

This arrangement relieves individual outlets from

holding large stocks and pools inventory risk

centrally. The central warehouse ships products

frequently based on demand, minimizing total

inventory while keeping shelves adequately stocked.

As a result, HLA has “reduced store inventory

pressure” and largely avoided overstocking at

individual outlets. Suppose a particular apparel style

sells poorly in one region. In that case, the company

can swiftly reroute those items to stores where

demand is higher rather than marking them down at

the original location.

HLA also employs centralized procurement. The

firm negotiates with its suppliers (who produce many

of its garments) on a consolidated, company-wide

basis, securing bulk discounts on materials and

manufacturing. This centralized procurement strategy

and HLA’s significant bargaining power have

enabled agreements whereby unsold products can be

returned to specific suppliers. In effect, HLA shifts

some inventory risk upstream to suppliers through

these buy-back arrangements, a practice facilitated by

the company’s scale and centralized negotiations.

This approach further reduces HLA’s inventory costs

and risk (Chen, 2022).

4.2.3 Technology and AI Utilization

After putting the structural foundations in place, HLA

recognized that technology was needed to manage its

enormous scale efficiently. In 2014, the company

began implementing Radio-Frequency Identification

(RFID) technology to tag and track apparel inventory

throughout its supply chain. HLA achieved real-time

visibility into inventory movements by tagging each

item and deploying RFID scanners at its central

warehouse and stores. This dramatically improved

inventory accuracy and reduced manual stock-taking.

According to internal analysis, inventory counting

time decreased by roughly 30% after RFID adoption

(Chen, 2022).

Building on enhanced data visibility, HLA

invested in AI-driven analytics to improve demand

forecasting and stock replenishment. It developed AI

models to predict fashion trends based on historical

sales, seasonality, and in-store customer traffic

patterns. It used these predictions to allocate

inventory from the central warehouse to stores. This

AI-driven forecasting allows HLA to respond swiftly

to fast-selling items (triggering rapid restocking) and

to detect slow-moving products (prompting

promotions or inter-store transfers). HLA’s

centralized IT system aggregates daily sales data from

all stores to feed these algorithms. As a result of these

optimizations, HLA reduced its inventory backlog by

approximately 30%, with a corresponding decrease in

inventory carrying costs. Inventory turnover reached

about 4.0 turns per year, compared to roughly 11.5

turns/year for fast-fashion leader Zara. Zara’s higher

turnover is driven by a highly responsive, semi-

decentralized model (shipping new products to stores

multiple times per week).

In contrast, HLA’s model prioritizes centralized

control with slightly longer cycles and strict cost

management. HLA’s use of AI is still evolving. The

company has recently explored AI for design (trend

analysis) and dynamic pricing, though those efforts

lie beyond our scope here. The key point is that HLA

exemplifies a data-enabled, centralized retail supply

chain focused on operational efficiency.

ICEML 2025 - International Conference on E-commerce and Modern Logistics

274

4.2.4 Performance Outcomes

HLA’s centralized, tech-driven supply chain strategy

has delivered significant competitive benefits. The

company rapidly rose to become China’s top

menswear brand by sales, indicating that its supply

chain reliably supports thousands of outlets and

consistent customer demand. Franchisees benefit

from the pooled inventory system through

significantly reduced stock risks, which likely

facilitated HLA’s rapid expansion of franchised

stores. The RFID and AI initiatives also improved

operational efficiency: annual inventory counts that

once took days were completed much faster, and

inventory accuracy improved, ensuring high product

availability. Inventory record accuracy increased

substantially after the RFID rollout (Chen, 2022).

Customer service levels remain high because the

central system swiftly refills stores as products sell.

The trade-off in HLA’s model is reduced flexibility:

it is less immediately responsive to new fashion

trends than Zara’s ultra-fast supply chain. HLA relies

on robust initial forecasts and mid-season

adjustments, so if those forecasts are off, there is a lag

before AI-driven corrections take effect. Nonetheless,

focusing on cost leadership, HLA’s centralized model

has successfully driven growth and efficiency.

However, unlike Unilever, HLA did not incorporate

explicit sustainability initiatives into its supply chain

strategy. Thus, it reaped the efficiency gains of

centralization and digitalization but did not achieve

the environmental improvements that a greener

approach could have provided – a gap that may pose

a strategic risk as sustainability becomes a more

prominent concern.

Both Unilever and HLA illustrate the

performance impact of combining centralized supply

chain structures with data-driven practices. However,

their scope and strategic priorities differ, most

notably in their incorporation of sustainability.

5 THEORETICAL AND

MANAGERIAL IMPLICATIONS

This research contributes to supply chain

management theory by demonstrating that integrating

AI-driven analytics, centralized structures, and green

practices yields a competitive capability that

competitors find difficult to replicate (Kamble &

Gunasekaran, 2019). It also shows that digitalization

can shift traditional contingency trade-offs: advanced

analytics enable a centralized model to remain agile

even under conditions such as high-demand volatility

that were once thought to require decentralization.

Furthermore, the evidence confirms that

environmental sustainability can go together with

profitability, supporting Porter's hypothesis that eco-

efficiency drives innovation and performance.

However, institutional context can act as a boundary

condition for these benefits: the multinational case

firm’s intense regulatory and stakeholder pressures

led to early adoption of GSCM, whereas the

emerging-market firm’s weaker external pressures

delayed such initiatives. The findings underscore the

importance of treating technology, structure, and

sustainability as an integrated socio-technical system.

Alignment across these domains is crucial since

focusing on one dimension in isolation may yield

short-term gains but not sustained, long-term

performance.

Practically, the results urge managers to pursue an

integrated strategy rather than isolated initiatives.

Case evidence indicates that investments in AI tools,

supply chain centralization, and GSCM deliver the

most significant value when implemented together in

a mutually reinforcing way. A strong data

infrastructure should underpin these efforts since

unified data makes AI applications far more effective.

Equally important is proactive change management;

senior leadership must communicate the benefits of

centralization and analytics to overcome resistance.

For example, Unilever secured local buy-in by

transparently sharing performance improvements and

involving local staff in centralized processes (Brito,

2016). Finally, managers should emphasize the

natural synergies between efficiency and

sustainability to build a compelling business case for

green initiatives. Many measures that reduce

environmental impact, such as optimizing routes or

packaging, also lower costs. Unilever’s logistics

program, for instance, cut carbon emissions by about

20% while saving approximately €91 million (Brito,

2016), illustrating how integrating AI, centralization,

and GSCM can achieve economic and environmental

gains (Qu & Kim, 2024).

6 CONCLUSION

This study employed a qualitative, case-based

methodology to examine how AI-driven analytics,

supply chain centralization, and green supply chain

management (GSCM) jointly enhance performance.

By comparing two firms-Unilever and HLA-the

research finds that an integrated strategy across

technological, structural, and environmental

dimensions yields synergistic improvements in

Path Analysis of Supply Chain Management Reform Driven by Data Empowerment

275

operational efficiency and environmental

sustainability. Unilever’s fully centralized, data-

driven, sustainability-focused model achieved

concurrent gains in cost, service levels, and carbon

footprint, whereas HLA’s omission of certain green

practices highlighted the missed opportunities when

integration is incomplete. These cases underscore that

AI and centralization empower agility and precision,

while GSCM ensures efficiency gains do not come at

the expense of ecological goals.

The findings extend supply chain management

theory by bridging strategic, technological, and

sustainability perspectives. They support the

resource-based view and dynamic capabilities

frameworks, showing that unique bundles of

capabilities (AI analytics, centralized structures, and

sustainability orientation) confer competitive

advantages that are hard to replicate.

Practically, this conclusion offers a roadmap for

practitioners. It demonstrates that pursuing AI,

centralization, and green initiatives in tandem-rather

than in isolation-can create complementary benefits.

Managers are advised to orchestrate digital

innovation, structural alignment, and environmental

responsibility holistically, breaking down silos

between efficiency and sustainability agendas to

achieve long-term resilience.

The insights are drawn from only two case

studies, which limits generalizability. It remains

challenging to disentangle the individual effect of

each component due to their integrated deployment,

and the rapid evolution of AI technologies means

today’s conclusions may require continual

revalidation. Future research should, therefore,

validate these findings across broader samples and

quantitative analyses and explore the underdeveloped

social sustainability dimension of AI-enabled

centralized supply chains. On this frontier, current AI

applications address environmental and economic

issues far more than social issues.

REFERENCES

Barney, J., 1991. The resource-based view of the firm.

ResearchGate.

Chen, X., 2022. Analysis of Supply Chain Cost

Management of HLA.

De, B., Magalhães, R., Brito, S., Rosário, M., De Oliveira,

M., & Moreira, A., 2016. Centralization of supply chain

management operations: the case of Unilever

Ultralogistik Dissertation Master in Management.

Eppen, G.D., 1979. Note—Effects of Centralization on

Expected Costs in a Multi-Location Newsboy Problem.

Management Science 25(5): 498–501.

Esfahbodi, A., Zhang, Y., Liu, Y., & Geng, D., 2023. The

fallacy of profitable green supply chains: The role of

green information systems (GIS) in attenuating the

sustainability trade-offs. International Journal of

Production Economics 255: 108703.

Feng, M., Yu, W., Wang, X., Wong, C.Y., Xu, M., & Xiao,

Z., 2017. Green supply chain management and financial

performance: The mediating roles of operational and

environmental performance. Business Strategy and the

Environment 27(7): 811–824.

Green, K.W., Zelbst, P.J., Meacham, J., & Bhadauria, V.S.,

2012. Green supply chain management practices:

impact on performance. Supply Chain Management: An

International Journal 17(3): 290–305.

Kalyar, M.N., Shafique, I., & Abid, A., 2019. Role of lean

manufacturing and environmental management

practices in eliciting environmental and financial

performance: the contingent effect of institutional

pressures. Environmental Science and Pollution

Research 26(24): 24967–24978.

Kamble, S.S., & Gunasekaran, A., 2019. Big data-driven

supply chain performance measurement system: a

review and framework for implementation.

International Journal of Production Research 58(1): 1–

22.

Mäenpää, M., 2024. Using artificial intelligence for data-

driven supply chain management: a global ICT

company case study. Oulu.fi.

Nguyen, D.T., Adulyasak, Y., Cordeau, J.-F., & Ponce, S.,

2022. Data-driven operations and supply chain

management: established research clusters.

Qu, C., & Kim, E., 2024. Reviewing the Roles of AI-

Integrated Technologies in Sustainable Supply Chain

Management: Research Propositions and a Framework

for Future Directions. Sustainability 16(14): 6186.

Walker, H., Di Sisto, L., & McBain, D., 2008. Drivers and

barriers to environmental supply chain management

practices: Lessons from the public and private sectors.

Journal of Purchasing and Supply Management 14(1):

69–85.

Zhu, Q., Sarkis, J., & Lai, K., 2013. Institutional-based

antecedents and performance outcomes of internal and

external green supply chain management practices.

Journal of Purchasing and Supply Management 19(2):

106–117.

ICEML 2025 - International Conference on E-commerce and Modern Logistics

276