Building Technological Capability in Indonesia’s Defense Industry: The Mediating Roles of Organizational Innovation and Technology Management Capability

The defense industry is widely recognized as a strategic arena underpinning deterrence, sovereignty, and national security. Within this sector, technological superiority has historically generated not only military advantages but also substantial spillovers into civilian industries, as innovations developed for defense applications have diffused into commercial products, production systems, and broader technological ecosystems. Contemporary defense operations increasingly integrate cyber-physical systems, artificial intelligence, and robotics to enhance efficiency, precision, and operational resilience, reinforcing the view that competitiveness in defense manufacturing and services is closely associated with advanced technology development and deployment [1, 2].

Simultaneously, the defense industry is structurally distinct from many other industrial sectors. National defense industries typically exhibit high capital intensity, long research and development (R&D) cycles, and strong economies of scale. These characteristics can make defense production commercially challenging, particularly in developing countries, where comparative advantage may lie in less capital-intensive sectors. However, the strategic imperative of defense autonomy often leads states to preserve or expand domestic defense industrial capacity through policy protection, subsidies, and procurement arrangements that stabilize demand. Consequently, defense industries operate within a hybrid governance environment shaped by security objectives, industrial development strategies, and market constraints, creating a complex institutional setting in which capability development is often as important as cost efficiency.

This context is directly relevant to Indonesia’s National Defense Industry (INHAN), which is being reshaped by rapid technological changes and intensifying capability requirements. Digital transformation and Industry 4.0 technologies are altering competitive and institutional conditions, creating opportunities for market expansion, broader stakeholder engagement, and improved operational efficiency, while simultaneously raising the performance thresholds that firms must meet to remain competitive. In increasingly dynamic competition, technology-based advantages can be disrupted quickly when capability renewal does not keep up. Consequently, transformation in INHAN should be understood as a strategic process of capability building and organizational renewal, rather than a narrow agenda focused on adopting tools, implementing automation, or adding isolated digital features.

Within this transformation trajectory, three internal capability domains become particularly salient: Absorptive Capacity (AC), Organizational Innovation (OI), and Technology Management Capability (TMC). Firms must develop AC to leverage external knowledge and translate it into productive internal routines; strengthen OI to reconfigure structures, processes, and decision rights so that learning can be converted into innovation outcomes; and build TMC to govern technology choices, align technological investments with strategy, and ensure that technology generates measurable organizational value [2-4]. Together, these capabilities form a practical basis for understanding why some firms respond to technological turbulence with upgrading and competitiveness, while others experience stagnation or disruption in their operations.

Industry 4.0 technologies are frequently positioned in the literature as levers for sustained competitive advantage, particularly in technology-intensive industries, where the speed of technological change and the importance of system integration shape competitive outcomes. Technological Capability (TC) can be conceptualized as the integration of technology across production and service operations, where cyber-physical systems and intelligent applications enable new production configurations and expand the feasibility of production data analytics [2]. Under this conceptualization, TC is not limited to owning machinery or software; it is expressed through the organization’s ability to integrate digital and physical systems, redesign work processes, collect and interpret data for operational improvement, and sustain continuous technological upgrading throughout the value chain.

However, transformation is rarely linear in the defense industry. Defense firms often delay capability development, particularly in digital systems and data-driven production, owing to concerns about data integrity, cybersecurity, and strategic confidentiality, as well as continued reliance on traditional managerial tools and legacy production routines. This hesitation is costly because TC increasingly functions as a foundational condition for efficiency, revenue growth, and innovation performance. Where TC is underdeveloped, firms may struggle to achieve stable quality, meet delivery targets, or respond quickly to evolving operational requirements. Therefore, readiness assessment, which evaluates whether an organization possesses baseline technological and organizational prerequisites, is critical before selecting transformation roadmaps and business models that can translate technology investments into measurable performance improvements.

In practice, INHAN firms face a set of operational and strategic paradoxes that complicate their capability-building efforts. They are expected to reduce costs while investing in AC, even though AC development is resource-intensive and often requires sustained spending on learning systems, training, knowledge governance, and collaboration mechanisms. They are expected to accelerate production while being constrained by limited OI, including rigid structures, conservative decision-making processes, and weak cross-functional coordination. They are expected to pursue efficiency and modernization while lacking sufficient TMC to design and implement technology strategies and business models that capture value from technology adoption [5-7]. These paradoxes imply that technology-related challenges in INHAN cannot be treated as isolated engineering problems; they are closely linked to knowledge governance, organizational design and managerial decision architectures.

Because the business model mediates how firms transform technology and knowledge into value, it becomes a central lens through which capability-building challenges can be diagnosed. Research highlights that business models help firms recalibrate their processes, practices, and operations to capture emerging opportunities and survive fast-moving and unpredictable competition [8]. In a defense setting, this mediation role can be even more pronounced: firms may have access to technology or external partners but still fail to translate such inputs into performance improvements if they do not have the internal managerial and organizational capabilities to integrate and exploit technology. In this sense, TC depends on an enabling configuration that links external knowledge acquisition, internal learning, organizational redesign, and technology governance into a coherent system.

Within Indonesia, INHAN is positioned as a cornerstone of defense self-reliance, supplying multi-domain capabilities and electronic systems while functioning as a strategic partner of the government [9]. Although domestic production capacity supports key Indonesian Armed Forces requirements, the sector’s TC is often assessed as mid-level [10]. This assessment reflects both achievements and constraints: domestic firms have developed significant production experience but remain challenged by limits in advanced system integration, high-end component mastery, and sustained innovation cycles. Frequently cited examples include PT Pindad’s support for land-based defense systems, PT Dirgantara Indonesia’s collaboration in aircraft and helicopter production and development, and PT PAL’s role in constructing maritime platforms, including large-scale warships, with private firms contributing to fast patrol craft [11]. Strengthening INHAN’s fundamentals is also associated with improvements in technology and manufacturing readiness, including testing capacity, from components to fully integrated systems. Key pathways include licensing, co-production, production modernization, joint development, and the enhancement of R&D facilities, supported by universities and research institutes to reinforce basic-to-applied research linkages [12].

A useful theoretical foundation for understanding capability differences across firms is the Resource-Based Theory (RBT), which argues that resources and capabilities that are valuable, rare, difficult to imitate, and non-substitutable underpin sustained competitive advantage [13]. This perspective shifts the strategic focus from external market pressures toward the inventory, development, and renewal of internal capabilities. This suggests that firms should begin strategic planning by identifying and strengthening the resources and capabilities that generate durable advantages rather than relying solely on environmental positioning. This is particularly relevant in the defense industries, where the market structure is often characterized by monopsony procurement and policy-driven demand, making internal capability differences a primary driver of performance and competitiveness. According to this logic, firms that systematically invest in learning, technology governance, and organizational renewal are more likely to achieve sustainable advantages, even when market conditions are constrained.

The Value, Rarity, Imitability and Organization (VRIO) framework operationalizes RBT by providing an internal analytic tool to evaluate whether resources meet the Valuable, Rare, Inimitable, and Organized criteria, helping to identify which capabilities can generate a durable advantage [14]. In the INHAN setting, VRIO provides a structured way to evaluate whether TC, AC, OI, and TMC can function as sources of sustained advantage. For example, TC may be valuable in improving efficiency and meeting performance requirements; it may be difficult to imitate when embedded in firm-specific processes; and it may generate a sustained advantage when the firm is organized to exploit it through effective governance and talent systems.

However, RBT alone can be insufficient when the environment is characterized by rapid technological changes, shifting operational requirements, and evolving threats. In such dynamic contexts, resource-based rents can erode as technologies diffuse, standards change, or new entrants disrupt traditional advantages. This is why a dynamic capabilities lens is frequently used alongside RBT to explain continuous adaptation and renewal in technology- and manufacturing-oriented organizations [15-17]. Dynamic capabilities emphasize an organization’s ability to sense opportunities and threats, seize opportunities through investment and strategic choice, and reconfigure resources and processes to maintain relevance in changing environments. In the defense industry, where technology trajectories can change quickly and procurement expectations can shift, dynamic capability is often a necessary complement to the resource-based perspective.

This theoretical framework clarifies why managerial capability related to technology governance is strategically significant. TMC is increasingly viewed as central because maintaining advantage cannot rely solely on operational efficiency and cost minimization. Countries and firms that lead technological progress are often those that can manage technological assets effectively, align technology investments with their strategy, and convert technical options into organizational value [18]. In this view, technology development is valuable insofar as it produces realizable organizational outcomes; therefore, the capability to evaluate, select, integrate, and exploit technology becomes critical, particularly in a sector where technology choices are expensive, path-dependent, and frequently subject to security constraints [19]. In this sense, TMC is not simply an administrative function; it is a strategic managerial capability that shapes TC development over time.

AC provides another foundational pillar for capability development. It is defined as the ability to recognize, assimilate, and apply external knowledge [20]. In defense contexts, foreign technology transfer is often a critical input; absorbing and re-innovating external knowledge can accelerate the modernization of weapons capabilities, and foreign inputs can constitute a substantial share of capability development in some settings [1]. The key implication is that technology acquisition alone is insufficient: without strong AC, technology transfer risks producing dependence rather than capability upgrading. Conversely, when AC is strong, external knowledge can be recombined with internal knowledge, creating conditions for domestic capability accumulation.

The literature links AC to dynamic capabilities and R&D investment, showing that AC expands knowledge and skill bases, improves future information assimilation, and ultimately supports innovation development and TC [5, 21-23]. This linkage is particularly relevant for INHAN because firms are often required to integrate external knowledge from technology partners, universities, research institutions, suppliers, and government agencies. Where AC is weak, learning remains fragmented and difficult to institutionalize. When it is strong, firms can create routines for scanning and acquiring knowledge, building internal competence, and converting knowledge into applied innovation and production improvement.

OI provides the internal mechanism that often determines whether knowledge absorption yields sustained upgrading. OI, supported by leadership, human resource management, knowledge management, creativity, and innovation management, is argued to strengthen innovation performance and contribute to higher TC [24, 25]. OI can take many forms, including redesigning work processes, changes in decision authority, new collaboration arrangements, and revised incentive structures that encourage experimentation and learning. In defense firms, OI can be particularly difficult because of hierarchical structures, confidentiality constraints, and conservative risk cultures. However, it can be decisive in enabling cross-functional integration, accelerating learning cycles, and supporting the adoption of advanced manufacturing and digital technologies.

Importantly, despite the rapid growth of AC research, definitional ambiguity, construct validation challenges, and inconsistent empirical findings persist, motivating more structured empirical testing, particularly to clarify mediation mechanisms [5, 26]. These inconsistencies matter because they suggest that AC does not automatically translate into TC; rather, the translation likely depends on internal mechanisms that shape how knowledge is integrated and exploited. OI and TMC can serve as mediators. OI can explain how absorbed knowledge is converted into renewed processes and structures, whereas TMC can explain how technology choices, governance routines, and strategic alignment shape the conversion of knowledge and technology into operational and competitive outcomes.

In response to these theoretical and empirical gaps, the present study examines the effect of AC on TC in Indonesian defense industry firms, specifying OI and TMC as mediators. This approach tests both direct and indirect pathways to explain prior inconsistencies and illuminate the internal mechanisms through which INHAN can strengthen TC. The conceptual contribution lies in treating TC as a product not only of technology investment but also of a broader capability system in which knowledge absorption, organizational renewal, and technology governance jointly determine whether firms can sustain competitiveness under rapid technological change.

The study covered 54 firms classified within INHAN. Firm selection was not based solely on the Wikipedia list [46]; it also considered representation, resource constraints, operational diversity, accessibility, and data availability to maintain methodological rigor and capture variation in TC and innovation across different firm types and scales. Primary data were collected through a structured questionnaire and supplemented by interviews with a subsample of respondents. Secondary data were used to support the assessment of TC, drawing on internal company documents and relevant literature and peer-reviewed sources aligned with the study variables.

The study population comprised leaders, managers, and senior employees within 54 defense firms. In line with the definition of population as the full set of units whose characteristics are estimated [47], the population was operationalized as supervisor-, manager-, and director-level personnel, totaling 162 potential respondents in the sample. These groups were targeted because they are assumed to have the most informed perspective on firm strategy, resource allocation, technology adoption, innovation practices, and performance, and are more likely to have access to sensitive organizational information than other employees.

Purposive sampling, a non-probability approach that selects respondents based on criteria aligned with the research objective [48], was used. The eligibility criteria emphasized leadership position (supervisor/manager/director), responsibility for operational oversight or strategic decision-making, and at least five years of work experience. Although the initial design targeted 162 respondents, data screening was applied to improve the precision. After filtering by position, the dataset was reduced to 153 responses. Responses completed in less than two minutes were removed because pre-test evidence suggested that typical completion required 7–15 minutes, leaving 128 responses. Further screening excluded questionnaires with identical answers across all items, resulting in a final sample of 116 respondents. Data collection was conducted primarily via Google Forms to improve efficiency, allowing both online and offline completion to support objectivity.

The research instrument was a questionnaire using a five-point semantic differential scale, which was treated as interval data. Consistent with survey research practices, the questionnaire was administered in a standardized order; this approach also aligns with the broader definition of questionnaires that can include structured interviews and online formats [49]. Constructs were operationalized through indicators measured by multiple items, but the analysis was conducted at the indicator level rather than at the individual-item level. Following Hair et al. [50], indicators were considered closer representations of the underlying constructs; therefore, this study used indicator mean scores to represent each indicator, consistent with recommendations for instruments with many items [47]. This approach was intended to simplify the analysis, improve measurement stability, reduce measurement error, mitigate item-level bias via aggregation, support interpretation within a first-order model, and present results more clearly.

Instrument quality was assessed through validity and reliability tests. Validity is defined as the extent to which an instrument measures what it is intended to measure [51]. Because the model used reflective indicators, convergent validity was evaluated through outer loadings, applying a threshold of 0.70 for the item-to-construct correlations [52]. The Average Variance Extracted (AVE) was also used, with 0.50 as the minimum threshold for adequate convergent validity [52], and indicators with loadings below 0.70 were treated as failing validity–reliability requirements. Reliability was defined as the consistency of measurement across conditions [49]. Reliability assessment relied on composite reliability and internal consistency reliability (Cronbach’s alpha), with alpha ideally above 0.70, while acknowledging that some perspectives treat values above 0.50 as minimally acceptable, albeit indicating low reliability [49].

Data analysis proceeded by treating measurement evaluation as a prerequisite for hypothesis testing, followed by structural model estimation using path analysis. The study adopted a first-order construct specification rather than a higher-order model to improve interpretability, align with the theoretical framing, reduce complexity, and lower the risks of overfitting and estimation error given the sample-size constraints. Hypothesis testing was conducted using Structural Equation Modeling (SEM) to support prediction and model testing, allowing the estimation of direct, indirect, and total effects among exogenous and endogenous variables [49]. SEM was positioned as an integrating factor analysis, regression, and path analysis, and was suitable for handling interactions, non-linearity, measurement error, correlated errors, and correlated latent variables measured with multiple indicators [53]. Mediation analysis was applied to clarify the mechanisms through which mediators explain the relationships between independent and dependent variables [47]. In PLS-SEM, SmartPLS algorithm and bootstrapping procedures provide direct effects, total and specific indirect effects, and total effects, enabling both single- and multiple-mediation tests [53].

The study employed variance-based, non-parametric PLS-SEM using SmartPLS because it does not require normality assumptions, supports models with many constructs and indicators, and is suitable for relatively small samples, such as the 116 respondents in this study. SmartPLS also relies on bootstrapping and is primarily prediction-oriented compared to covariance-based SEM, which is more confirmatory and typically assumes multivariate normality. The explanatory power of the inner model was assessed using R² values. As an initial step, a pre-test with 38 respondents was conducted to assess the instrument strength. Given the limited pre-test sample, SPSS 26 was used for validity testing via Pearson correlations and reliability testing via Cronbach’s alpha [47]. The pre-test indicated that all indicators were significant at p < 0.05, and therefore valid. For the main sample size of 116, the critical r value at the 0.05 level was 0.183, and all the indicator correlations exceeded this threshold. Reliability was also supported, with Cronbach’s alpha values above 0.70 for all variables, as reported in the SPSS output.