Purpose of this paper:

In recent years, the structure of global industrial supply chains has undergone significant changes due to the impact of the US-China trade war and the COVID-19 pandemic, consequently elevating risks across the entire supply chain. Facing these shifting circumstances, Taiwan's industries find that past business models are no longer sufficient to address potential changes and risks. Only by assessing their own supply chain structural condition—specifically their capabilities and tolerance levels across various aspects—and actively responding to challenges can they maintain their competitiveness in a volatile global environment. To understand the supply chain resilience (SCR) of Taiwan's industries in the face of supply chain risks, the Ministry of Economic Affairs (MOEA) commissioned the Industrial Technology Research Institute (ITRI) to design a questionnaire on SCR and to investigate it across various industries in Taiwan. Over the past three years, we have been studying the SCR of industries such as semiconductors, machinery, textiles, microelectronics, etc. For each of those industries, we first invited 6-12 companies to fill out the questionnaire via https://drscr.org.tw/. After analysis, we further requested 5-6 companies, under their agreement, to conduct a physical visit to clarify and discuss the questionnaire results. We concluded with suggestions for improving supply chain resilience. This paper presents the questionnaire and discussion results for two textile companies, focusing on the supply chain strategies and actions they plan to implement and have already implemented, and on how they could further improve to respond to both short- and long-term risks and uncertainties, including responses to geopolitical turbulence.

Design/methodology/approach STS:

The questionnaire was initiated and designed by the MOEA of Taiwan to evaluate the SCR of Taiwan's industries as a reference for formulating industrial development and support programs to strengthen Taiwan's overall resilience. MOEA assigned the questionnaire development project to ITRI, which invited professors and industrial professionals to collaborate. After a year, the project team, using the Delphi Method, delivered the questionnaire design and analysis. To assess overall national resilience, WEF (2012) proposed 5R: Robustness, Redundancy, Resourcefulness, Response and Recovery as the lens of looking at five national-level subsystems (economic, environmental, governance, infrastructure and social). The 5R is one of the fundamental models used to develop our SCR model. Moreover, since the questionnaire is on SCR, in the development stage, the flows of goods along supply chains are also considered as 5 F(lows), starting from (1) product development, (2) demand management, (3) procurement management, to (4) production and (5) order fulfilment. After three rounds of collective critique on questionnaires with experts from the ITRI, industry and academia, the questionnaire for evaluating the SCR of a company was developed. However, the 5R represents a company's implicit capacity, but most of the time it is hard to associate them with managerial and strategic improvement. Thus, ITRI further classified the questionnaire variables into 5 explicit capabilities (5C) as Digital, Analytical, Decision-making, Collaborative, and Sustainability. These capabilities are directly linked to a business's supply chain practices and will primarily be used to represent a company's SCR in the later analysis. In total, there are 144 variables (questions) and they are divided into five constructs (capabilities). Each construct has 1-5 KSF (key success factors) and each KSF has 1-3 KPI (key performance index), while in each KPI, there are 1-15 variables. After filling out the questionnaire, the scores of 5Cs, 5Rs, 5Fs and their respective KPIs and KSFs can be calculated. These scores are very useful for analyzing the strengths and weaknesses of a company’s SCR.

In the past three years, diverse industries were selected for the study. In this paper, we show the results of the questionnaire and physical visits. The representative companies in the industry were selected and invited based on the suggestions from the industry association and professional experts. Most selected ones are publicly listed. After completing the questionnaires, 5-6 companies in the industry are selected again for physical visits. During the physical visit, we were received by high-ranking managers, usually from 2-4 divisions of the company, who instructed their staff to complete the questionnaire.

Findings:

We first display the questionnaire analysis results and then discuss with the company one by one what the results mean to them practically and how they could and plan to improve the SCR. Below is a partial list of the SCR of Company A. In addition to the 5 Capabilities of the SCR for a business, we will present the SCR from the perspectives of 5Rs and 5Fs, and show their relationships.

Value:

This paper examines the SCR of businesses and presents the variables, KPIs, KSFs, and questionnaire constructs developed through the Delphi Method after three rounds of questionnaire consensus involving numerous professionals and professors. In this study, we show the SCR for 5 Capabilities because they are explicit measurements and serve as an illustration of the company's SCR. Other resilience measurements, like 5Rs and 5Fs, can be easily converted to show the implicit measurement (5Rs) or the product flow (5Fs) of the SCR. Two physical visits to the case companies further provide practical meaning to the questionnaire results and how companies may, as a reference, respond to the volatile supply chain environment and geopolitical tension.

Research limitations/implications:

The questionnaire was developed with support from the MOEA of Taiwan to measure the industry's SCR. Despite following strict academic standards in developing the questionnaire, it is still influenced by the opinions of governmental officers, which reflects its utility for policy-making. The selected samples are mainly recommended by respective industry associations, with some recommendations from the MOEA, rather than a random selection process. Thus, the result may not reflect the industry's current status quo.

Practical implications:

The study's research limitations reflect its practical implications, as governmental officers and industrial professionals were involved in both the questionnaire development and the company selection process. The results, though uncomprehensive, show the SCR of two representative companies and how they respond to the volatilities and risks of the current supply chain environment.

Literature

Vargas, J. and González, D. (2016). Model to assess supply chain resilience. International Journal of Safety and Security Engineering, 6(2), 282–292.

World Economic Forum. (2013). Global Risks Report 2013: Eighth Edition. An Initiative of the Risk Response Network. Geneva: World Economic Forum.

World Economic Forum. (2025). Global Risks Report 2013: 20th Edition. A World of Growing Division. Geneva: World Economic Forum.

Yang, C., Tian, K. and Gao, X. (2025). Supply chain resilience: Measure, risk assessment and strategies, Fundamental Research, 5(2), 433-436.

1 Purpose of this Paper

Modern supply chains face unprecedented challenges from escalating short-term disruptions—natural

disasters, cyber-attacks, pandemics—and long-term systemic risks including geopolitical turbu-

lence, climate change, and demographic shifts [1]. The COVID-19 pandemic illustrated how

systems optimized solely for efficiency catastrophically fail when confronted with demand surges

and supply disruptions [3]. Traditional supply chain management approaches optimize for cost

minimization, inventory reduction, and capacity utilization under assumptions of stability, prov-

ing fundamentally inadequate when stability breaks down [4].

This paper addresses a critical gap: while existing research recognizes the importance of sup-

ply chain resilience, most frameworks remain sector-specific or provide only aggregate resilience

scores, limiting practical utility for managers who need granular insights into which capabilities

to develop, where to invest resources, and how to prioritize improvement initiatives [2]. Further-

more, existing resource allocation approaches typically address single resource categories and

focus on normal operations rather than crisis response [5].

The primary purpose is to introduce a paradigm shift toward resilience-driven supply chain

management where adaptive capacity takes precedence over static optimization. We develop a

comprehensive hierarchical control framework applicable across diverse supply chain contexts,

with particular focus on the first level—dynamic resource reallocation through internal mutu-

alization—demonstrating how intelligent coordination of existing resources can simultaneously

improve both crisis resilience and normal-operations efficiency without additional capital invest-

ment.

2 Design/Methodology/Approach

This research employs a multi-method approach combining mathematical modeling, algorithm

development, and simulation-based validation. We propose a three-level hierarchical control

framework comprising sequential intervention levers operating at different timescales: (1) dy-

namic resource reallocation (tactical, hours-to-days), (2) operational replanning (operational,

days-to-weeks), and (3) topological reconfiguration (strategic, months-to-years).

Focusing on Level 1, we develop three interconnected methodological contributions. First, we

construct a function-based performance evaluation model using graph theory where supply chain

networks are represented as directed graphs with nodes (facilities) and arcs (resource flows)

[6]. Functions are defined as integrated services requiring coordinated combinations of three

resource categories: equipment, medications/materials, and staff. Performance metrics employ

conjunctive logic—missing one critical input prevents entire function execution. Second, we

formalize resilience through a parametric mathematical model using hyperbolic tangent functions

with seven independent parameters capturing degradation and recovery dynamics, from which

six derived indicators enable multi-dimensional diagnosis [7]. Third, we develop priority-based

greedy allocation algorithms with complexity O(|N |2 × |T | × 3) suitable for real-time crisis

management.

Validation employs discrete-event simulation on a healthcare supply chain network comprising

three facilities over 24 time periods, with multiple crisis episodes simulating realistic disrup-

tion patterns [8]. Healthcare serves as validation context because it exemplifies broader supply

chain challenges: multiple interdependent resources, service-oriented operations, demand uncer-

tainty, capacity constraints, and catastrophic consequences of failure [3]. Comparative scenario

analysis quantifies performance differences between independent operation (Base scenario) and

coordinated resource sharing (Flow scenario).

3 Findings

Simulation results demonstrate that intelligent resource mutualization achieves performance

improvements of 40 percentage points during crisis periods without additional capacity invest-

ment [8]. Detailed temporal analysis reveals three critical advantages of coordinated resource

sharing.

First, recovery speed: the Flow scenario recovers to 90% performance within 3-5 periods after

initial shock, while the Base scenario struggles to exceed 50% performance even after 10 peri-

ods, representing 2x faster recovery. Second, sustained resilience: the Flow scenario maintains

near-perfect performance (90-100%) for extended stable periods, demonstrating that intelligent

coordination enables crisis recovery and sustains high operational efficiency [6]. Third, crisis ro-

bustness: during subsequent disruptions, the Flow scenario maintains 20-30% functionality while

the Base scenario collapses to below 10-15% performance, representing a 200-300% performance

advantage.

Analysis reveals that spatial heterogeneity—where some nodes operate at high capacity while

others remain underutilized—creates the fundamental opportunity for mutualization gains. Re-

source mutualization transforms network behavior from fragile systems exhibiting catastrophic

cascading failures to resilient networks capable of rapid recovery and sustained high performance

despite repeated shocks. These findings hold across diverse supply chain contexts including man-

ufacturing, distribution, humanitarian logistics, and service operations, as the mathematical

framework remains unchanged while adaptation lies in parameter specification [2].

4 Value

This research makes three original contributions advancing supply chain resilience theory and

practice. First, it introduces a novel function-based performance framework measuring system

capability to deliver integrated services requiring coordinated resource combinations—shifting

focus from resource-level metrics to value-creation capabilities [6]. Second, it develops a para-

metric resilience assessment approach with seven independent parameters and six derived indica-

tors enabling precise diagnosis of specific organizational weaknesses and targeted interventions,

addressing the limitation of existing aggregate resilience scores [2, 7]. Third, it demonstrates

that coordination capabilities often yield greater resilience improvements than equivalent invest-

ments in additional capacity—orders of magnitude in some cases—opening new pathways for

cost-effective supply chain improvement [4]. The framework’s generalizability across manufac-

turing, distribution, humanitarian, and service sectors establishes both theoretical foundations

and practical tools for implementing resilience-driven management [8].

5 Research Limitations/Implications

5.1 Research Limitations

This study focuses exclusively on Level 1 (dynamic resource reallocation) of the three-level hi-

erarchical framework, while complete resilience management requires integration of all three

levels. The validation employs simulation on a representative but simplified healthcare network;

real-world implementation would face additional organizational, regulatory, and behavioral com-

plexities [3]. The greedy allocation algorithm, while computationally efficient for real-time appli-

cation, does not guarantee global optimality—exact optimization approaches may yield marginal

improvements [5]. The framework assumes organizational willingness to share resources, which

requires establishing trust, governance structures, and incentive alignment mechanisms.

5.2 Implications for Future Research

Several extensions merit investigation. First, formalizing Levels 2 and 3 would complete the

framework, enabling integrated multi-level strategies. Second, incorporating exact optimiza-

tion solvers could benefit strategic planning contexts [5]. Third, extending the framework to

stochastic settings would capture demand/supply uncertainty and disruption unpredictabil-

ity [1]. Fourth, empirical validation across manufacturing, retail, and humanitarian contexts

would confirm generalizability beyond healthcare [4]. Fifth, investigating hybrid approaches

combining resource mutualization with other resilience strategies could enable comprehensive

resilience management [2]. Finally, behavioral research examining conditions under which orga-

nizations successfully implement resource sharing agreements would enhance practical applica-

bility.

6 Practical Implications

Results suggest actionable insights for supply chain practitioners and policymakers:

Invest in Coordination, Not Just Capacity: Intelligent coordination of existing resources

often yields greater resilience than capacity expansion at dramatically lower cost [8]. Organiza-

tions should prioritize developing coordination capabilities—information systems, communica-

tion protocols, relationship management—rather than solely investing in redundant capacity [4].

Design Flexible Resource Sharing Policies: Organizational policies governing resource

sharing significantly impact resilience independent of total resource inventories [6]. Policy re-

views examining barriers to internal resource mobility may represent the most cost-effective

resilience interventions. Organizations should establish clear governance frameworks defining

when, how, and under what conditions resources can be shared.

Activate Mutualization Dynamically: Resource mutualization provides minimal normal-

operations benefit but critical crisis value. Rather than maintaining permanent coordination

overhead, organizations should design rapid activation mechanisms—pre-established protocols,

decision authority, communication channels—that can be triggered when disruptions occur [7].

Establish Cross-Organizational Mutual Aid Frameworks: The principles extend beyond

internal organizational boundaries. Industry associations, regional networks, and governmental

bodies can facilitate mutual aid agreements enabling resource sharing across organizations during

crises [2].

Address Geopolitical and Systemic Risks: The framework directly addresses both short-

term disruptions and long-term systemic risks including geopolitical turbulence [1]. Dynamic

reallocation provides tactical response to immediate disruptions while supporting strategic re-

silience against long-term uncertainties.

References

[1] R. Agrawal et al., Progress and trends in integrating Industry 4.0 within Circular Economy,

Business Strategy and the Environment, vol. 33, no. 1, pp. 62–81, 2024.

[2] D. Ivanov, The Industry 5.0 framework, International Journal of Production Research, vol.

61, no. 5, pp. 1683–1695, 2023.

[3] L. Keshtkar et al., Resilience of healthcare supply chains, International Journal of Production

Research, vol. 62, no. 8, pp. 2918–2950, 2024.

[4] S. K. Paul et al., Key enablers of resilient and sustainable construction supply chains, Sus-

tainable Production and Consumption, vol. 36, pp. 323–338, 2023.

[5] V. Salehi et al., Modeling and analysis of oil and gas projects’ expenditure, Journal of Cleaner

Production, vol. 356, 131849, 2022.

[6] N. Rahiel et al., Function-based modeling for reactive optimization, in Proc. LOGISTIQUA

2025 (IEEE), Casablanca, Morocco, 2025.

[7] N. Rahiel et al., Mathematical modeling of supply chain resilience, in Proc. IN4PL 2025

(IFAC), Marbella, Spain, 2025.

[8] N. Rahiel, Formalisation mathématique de la résilience, Ph.D. dissertation, Université Paris

8, France, 2025.