Product disassembly plays a crucial role in the recycling, remanufacturing, and reuse of end-of-use (EoU) products. However, the current manual disassembly process is inefficient due to the complexity and variation of EoU products. While fully automating disassembly is not economically viable given the intricate nature of the task, there is potential in using human–robot collaboration (HRC) to enhance disassembly operations. HRC combines the flexibility and problem-solving abilities of humans with the precise repetition and handling of unsafe tasks by robots. Nevertheless, numerous challenges persist in technology, human workers, and remanufacturing work, which require comprehensive multidisciplinary research to address critical gaps. These challenges have motivated the authors to provide a detailed discussion on the opportunities and obstacles associated with introducing HRC to disassembly. In this regard, the authors have conducted a review of the recent progress in HRC disassembly and present the insights gained from this analysis from three distinct perspectives: technology, workers, and work.
This content will become publicly available on August 28, 2025
A Disassembly Score for Human-Robot Collaboration Considering Robot’s Capabilities
Product disassembly is essential for remanufacturing operations and recovery of end-of-use devices. However, disassembly has often been performed manually with significant safety issues for human workers. Recently, human-robot collaboration has become popular to reduce the human workload and handle hazardous materials. However, due to the current limitations of robots, they are not fully capable of performing every disassembly task. It is critical to determine whether a robot can accomplish a specific disassembly task. This study develops a disassembly score which represents how easy is to disassemble a component by robots, considering the attributes of the component along with the robotic capability. Five factors, including component weight, shape, size, accessibility, and positioning, are considered when developing the disassembly score. Further, the relationship between the five factors and robotic capabilities, such as grabbing and placing, is discussed. The MaxViT (Multi-Axis Vision Transformer) model is used to determine component sizes through image processing of the XPS 8700 desktop, demonstrating the potential for automating disassembly score generation. Moreover, the proposed disassembly score is discussed in terms of determining the appropriate work setting for disassembly operations, under three main categories: human-robot collaboration (HRC), semi-HRC, and worker-only settings. A framework for calculating disassembly time, considering human-robot collaboration, is also proposed.
more »
« less
- Award ID(s):
- 2026276
- PAR ID:
- 10544010
- Publisher / Repository:
- Proceedings of the ASME 2024 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, IDETC/CIE2024, August 25–28, 2024, Washington, DC, USA.
- Date Published:
- Format(s):
- Medium: X
- Location:
- Washington, DC, USA
- Sponsoring Org:
- National Science Foundation
More Like this
-
more » « less
Abstract -
Disassembly is an integral part of maintenance, upgrade, and remanufacturing operations to recover end-of-use products. Optimization of disassembly sequences and the capability of robotic technology are crucial for managing the resource-intensive nature of dismantling operations. This study proposes an optimization framework for disassembly sequence planning under uncertainty considering human-robot collaboration. The proposed model combines three attributes: disassembly cost, disassembleability, and safety, to find the optimal path for dismantling a product and assigning each disassembly operation among humans and robots. The multi-attribute utility function has been employed to address uncertainty and make a tradeoff among multiple attributes. The disassembly time reflects the cost of disassembly and is assumed to be an uncertain parameter with a Beta probability density function; the disassembleability evaluates the feasibility of conducting operations by robot; finally, the safety index ensures the safety of human workers in the work environment. The optimization model identifies the best disassembly sequence and makes tradeoffs among multi-attributes. An example of a computer desktop illustrates how the proposed model works. The model identifies the optimal disassembly sequence with less disassembly cost, high disassembleability, and increased safety index while allocating disassembly operations between human and robot. A sensitivity analysis is conducted to show the model's performance when changing the disassembly cost for the robot.more » « less
-
Abstract Disassembly is an essential step for remanufacturing end-of-life (EOL) products. Optimization of disassembly sequences and the utilization of robotic technology could alleviate the labor-intensive nature of dismantling operations. This study proposes an optimization framework for disassembly sequence planning under uncertainty considering human–robot collaboration. The proposed framework combines three attributes: disassembly cost, safety, and complexity of disassembly, namely disassembleability, to identify the optimal disassembly path and allocate operations between human and robot. A multi-attribute utility function is used to address uncertainty and make a tradeoff among multiple attributes. The disassembly time reflects the cost of disassembly which is assumed to be an uncertain parameter with a Beta distribution; the disassembleability evaluates the feasibility of conducting operations by robot; finally, the safety index ensures the protection of human workers in the work environment. An example of dismantling a desktop computer is used to show the application. The model identifies the optimal disassembly sequence with less disassembly cost, high disassembleability, and increased safety index while allocating disassembly operations among human and robot. A sensitivity analysis is conducted to show the model's performance when changing the disassembly cost for the robot.more » « less
-
Human-robot collaboration (HRC) has become an integral element of many industries, including manufacturing. A fundamental requirement for safe HRC is to understand and predict human intentions and trajectories, especially when humans and robots operate in close proximity. However, predicting both human intention and trajectory components simultaneously remains a research gap. In this paper, we have developed a multi-task learning (MTL) framework designed for HRC, which processes motion data from both human and robot trajectories. The first task predicts human trajectories, focusing on reconstructing the motion sequences. The second task employs supervised learning, specifically a Support Vector Machine (SVM), to predict human intention based on the latent representation. In addition, an unsupervised learning method, Hidden Markov Model (HMM), is utilized for human intention prediction that offers a different approach to decoding the latent features. The proposed framework uses MTL to understand human behavior in complex manufacturing environments. The novelty of the work includes the use of a latent representation to capture temporal dynamics in human motion sequences and a comparative analysis of various encoder architectures. We validate our framework through a case study focused on a HRC disassembly desktop task. The findings confirm the system's capability to accurately predict both human intentions and trajectories.more » « less
-
more » « less
Abstract Human-robot collaboration (HRC) has become an integral element of many industries, including manufacturing. A fundamental requirement for safe HRC is to understand and predict human intentions and trajectories, especially when humans and robots operate in close proximity. However, predicting both human intention and trajectory components simultaneously remains a research gap. In this paper, we have developed a multi-task learning (MTL) framework designed for HRC, which processes motion data from both human and robot trajectories. The first task predicts human trajectories, focusing on reconstructing the motion sequences. The second task employs supervised learning, specifically a Support Vector Machine (SVM), to predict human intention based on the latent representation. In addition, an unsupervised learning method, Hidden Markov Model (HMM), is utilized for human intention prediction that offers a different approach to decoding the latent features. The proposed framework uses MTL to understand human behavior in complex manufacturing environments. The novelty of the work includes the use of a latent representation to capture temporal dynamics in human motion sequences and a comparative analysis of various encoder architectures. We validate our framework through a case study focused on a HRC disassembly desktop task. The findings confirm the system’s capability to accurately predict both human intentions and trajectories.
