主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

• •    

数据资源整合“三层次”维度如何影响制造企业服务化?——基于“一主二辅”效应逻辑 (中国“双法”研究会管理决策与信息系统2025年会高质量优秀论文)

叶宝升, 户安涛, 张振刚   

  1. 福建农林大学经济与管理学院, 350002
  • 收稿日期:2025-06-10 修回日期:2026-04-29 接受日期:2026-06-05
  • 通讯作者: 张振刚
  • 基金资助:
    福建省社科基地重大项目“新质生产力背景下福建绿色产品价值共创研究”(FJ2024JDZ035); 国家社会科学基金重大项目“数据赋能激励制造业企业创新驱动发展及其对策研究”(18ZDA062)

How Do the Multiple Dimensions of Data Resource Integration Influence Servitization in Manufacturing Enterprises?—A Logic of “One Primary and Two Secondary Effects”

  1. , 350002,
  • Received:2025-06-10 Revised:2026-04-29 Accepted:2026-06-05

摘要: 如何释放数据资源价值以赋能制造企业服务化,是一个重要但尚未被充分关注的问题。现有研究多从技术或组织视角切入,将数据资源视为数字技术与平台的“被动”产物,忽视了企业在数据资源整合过程中的战略主动性。基于数据资源整合的“思想—活动—支撑”三层次维度,本研究提出“一主二辅”效应逻辑,首先探究活动层的科学型、市场型与供应链型数据资源识别获取如何通过稳定调整型、开拓创造型数据资源转化利用促进制造企业服务化(“一主”效应),而后分析思想层、支撑层与活动层要素联合影响制造企业服务化的条件组态(“二辅”效应)。通过PLS-SEM与fsQCA的混合方法,实证结果表明:(1)科学型数据资源识别获取主要通过开拓创造型数据资源转化利用提升服务化,而稳定调整型数据资源转化利用表现出非线性中介特征。(2)市场型数据资源识别获取通过稳定调整型和开拓创造型数据资源转化利用两种方式均可提升服务化。(3)供应链型数据资源识别获取主要通过稳定调整型数据资源转化利用提升服务化,而开拓创造型数据资源转化利用表现出非线性中介关系。(4)实现高水平制造企业服务化的条件组态有3种类型:“6条件全部存在”型、“6条件存在+1条件缺失”型、“7条件全部存在”型。研究发现不仅深化数据资源作为战略要素的价值实现机制,也拓展制造企业服务化前因变量体系。

关键词: 数据资源, 制造企业服务化, 数据资源识别获取, 数据资源转化利用

Abstract: Unlocking the value of data elements remains an important yet underexplored issue. Existing research on promoting servitization in manufacturing enterprises has largely focused on “digital technologies,” often overlooking the critical role of “data resources.” This study, based on a three-dimensional framework of data resource integration, proposes a “one primary and two secondary effects” logic. It investigates how the identification and acquisition of scientific, market, and supply chain data resources drive servitization through the stable-adjustment and exploratory-creative transformation of data resources (the “primary effect). Additionally, it explores how factors at the ideological, supportive, and operational levels combine to influence the conditions for servitization (the “secondary effects”). Using a mixed-method approach of PLS-SEM and fsQCA, the empirical findings reveal the following: (1) Scientific-type data resource acquisition enhance servitization through pioneering-type data resource utilization, but a nonlinear mediation effect exists when using stabilizing-type utilization. (2) Market data resource acquisition boost servitization via both stabilizing-type and pioneering-type data resource utilization. (3) Supply chain data resource acquisition promote servitization through stabilizing-type data resource utilization, though a nonlinear mediation effect is found with pioneering-type data resource utilization. (4) Three distinct configurations lead to high-level servitization: “all six conditions present,” “six conditions present with one missing,” and “all seven conditions present.” These findings contribute to the literature on both the antecedents of servitization in manufacturing enterprises and the realization of data resource value.

Key words: Data resources, Servitization, Data resource acquisition, Data resource utilization