在全球化和数字化转型的背景下，协同办公在现代组织管理中发挥着重要作用，远程工作模式成为新常态，飞书和钉钉等协同办公平台已成为提升组织效率和促进组织协同的关键工具。尽管如此，用户在使用这些协同办公工具时仍面临信息过载、任务协调困难以及沟通效率低下等一系列挑战。人工智能尤其是大语言模型的飞速发展，为解决这些问题提供了新的契机，从而提升工作效率和体验。本研究的目的是通过深入探索协同办公场景下的用户需求与痛点，结合先进的人工智能技术，设计出基于降低认知负荷理论的人工智能（AI）助手功能，以提升工作效率和办公体验。

本研究采用质性研究方法，对10名具有丰富协同办公经验的用户进行了深度访谈，并依据扎根理论进行编码分析，识别出七个核心范畴及其下属的23个二级编码和270个一级编码。这些编码详细描述了用户在协同办公场景中的具体问题和需求，包括日程和项目管理的困难、会议和沟通效率低下以及数据分析等难题，和高效日程安排、快速沟通、便捷发起会议、以及高效文档输出等方面的需求。研究基于这些成果，绘制了用户画像和旅程图，梳理出19个具体需求并转化为AI助手的18个功能点，基于降低认知负荷理论进行了功能的设计。最终产出了功能的高保真原型，并在可用性测试中得分83.75，评级“A”。最后，根据测试结果和用户反馈进行了产品迭代设计。

总体而言，本研究的设计成果可优化用户在协同办公场景中的工作体验，有效解决用户的痛点，提升工作效率。并且结合AI技术和降低认知负荷理论拓宽了该场景解决方案的新思路。

In the context of globalization and digital transformation, collaborative work plays a crucial role in modern organizational management, with remote work becoming the new norm. Platforms such as Feishu and DingTalk have become key tools for enhancing organizational efficiency and promoting collaboration. However, users still face challenges such as information overload, task coordination difficulties, and low communication efficiency when using these collaborative tools. The rapid development of artificial intelligence, particularly large language models, offers opportunities to address these issues and improve work efficiency and experience. This study aims to explore the user needs and pain points in collaborative work scenarios, combining advanced AI technology to design AI assistant functions based on cognitive load reduction theory to enhance work efficiency and office experience.

This study employs qualitative research methods, conducting in-depth interviews with 10 users experienced in collaborative work. Grounded theory was used for coding analysis, identifying seven core categories, 23 secondary codes, and 270 primary codes. These codes detail specific issues and needs in collaborative work scenarios, including difficulties in schedule and project management, low meeting and communication efficiency, and data analysis challenges, as well as needs for efficient scheduling, quick communication, easy meeting initiation, and efficient document output. Based on these findings, user personas and journey maps were created, identifying 19 specific needs and translating them into 18 AI assistant function points, optimized based on cognitive load reduction theory. A high-fidelity prototype of the functions was produced, scoring 83.75 in usability tests with an "A" rating. Finally, product iteration design was conducted based on test results and user feedback.

Overall, the design outcomes of this study optimize the user experience in collaborative work scenarios, effectively addressing user pain points and significantly improving work efficiency. By integrating AI technology with cognitive load reduction theory, this study expands new solution approaches for this context.