Objective The practice and research in the field of landscape architecture are currently characterized by a certain degree of separation. However, with the continuous expansion of the field of design practice, design boundaries are becoming increasingly blurred, and modern landscape design is facing an imminent process of scientific advancement. In order to incorporate landscape architecture into the academic system as a disciplinary field, there is a pressing need within the landscape architecture community to emphasize the importance of returning to the essence of design, as well as to address the severity of the gap between academic research and design practice. To achieve this, the landscape architecture community must actively establish academic autonomy, collectively cultivate a core methodology for design research that aligns with its own cognitive characteristics, and ensure a logically adaptive design research paradigm.

Methods/process  To address the pressing needs at present, this research, through literature review, conceptual analysis, and theory building, summarizes three main types of traditional design research, including Research about Design (RaD), Research for Design (RfD), and Research through Design (RtD), and determines the positional relationships between their design and research components. In terms of design practice, the research explores the stages of the design process through discussions on domestic and international design steps, and accordingly establishes the three stages of pre-design, design, and post-design, with the pre-design stage focusing on information gathering and problem analysis, the design stage focusing on design progression and solution evaluation, and the post-design stage focusing on design reflection and post-construction evaluation. Based on the previously mentioned relationships between design and research in RaD, RfD, and RtD, the research reveals a coupling relationship between these three approaches and the aforesaid three stages of the design process. RfD, which primarily involves preliminary investigation and research, is positioned in the pre-design stage; RaD, which mainly interprets the design outcomes, is situated in the post-design stage; RtD, which integrates design and research extensively, is positioned in the design stage.

Results/conclusion  In summary, a reconstructed landscape design research system is proposed, with design process as the main driver and “dual parallel lines and three loops” as the main characteristic. The “dual parallel lines” represent the intertwined design-based research and research-based design. Design-based research refers to research based on design, guiding the methodology of design research. This type of research focuses on design products, designers, or the design process. It refines scientific questions from design problems with a focus on problem solving, thus gradually generating the optimal solution. It includes the three stages of RfD, RtD, and RaD, which are respectively corresponding to deductive/quantitative research, mixed/retrospective research, and inductive/qualitative research. Research-based design refers to design based on research, guiding the practical application of design research. This type of design aims to generate design solutions to address design problems. It includes three stages: Feedforward design, process iteration design, and feedback design. The ultimate goal is to achieve a “satisfactory solution” or “solution set”, thus producing outstanding design outcomes or solutions with the potential to develop new knowledge. The two parallel lines of activity follow separate paths, both traversing the three stages of pre-design, design, and post-design, forming the “three loops” section. Based on differences in their respective focuses, each stage is further differentiated into research outcomes and design outcomes. The outcomes of the pre-design stage are generated by RfD and feedforward design types, those of the design stage are produced by RtD and process iteration design types, and those of the post-design stage are delivered by RaD and feedback design types. Based on the three stages of pre-design, design, and post-design, this research provides corresponding examples to illustrate each part of the “three loops”. It attempts to demonstrate how traditional design research types can be extended and developed to integrate with design practices, or how design practices can adopt comprehensive and interdisciplinary scientific approaches to explore new knowledge, complete iterative solutions, and solve design problems. Throughout the demonstration process, this research precisely defines the design practices and scientific research types at each loop. By highlighting the differences between design thinking and scientific thinking, the research explains how cyclic interdependence is established between design processes and research activities in each of the three stages. The research also elaborates on how different approaches can be chosen to conduct design-oriented research or research-oriented design, thereby developing satisfactory design outcomes and research outcomes. In conclusion, the construction of a landscape design research system led by the design process is beneficial for researchers and designers in different stages of the design process to break free from inherent perspectives, and can help them understand the three relationships between design and research in an academic context, consciously incubate design outcomes and research outcomes, and bridge the gap between design and research. This research hopes to further advance the current field of design research in landscape architecture, and clarify the future development path of landscape design.