The construction job sites are highly dynamic and chaotic environment for both workers and management. To ensure efficient task performances, workspace is a key resource on construction sites. To contribute to enhancing lean construction approaches with attempts to obtain real and objective data, our research team has developed a real-time positioning system based on Bluetooth Low Energy (BLE) technology where coordinates of workers can be obtained, and heat maps of task-related workspace are generated in real time. Our research team aims to develop an automated process of generation of workspace based on coordinates of workers and to explore various uses of workspace heat maps to serve the purposes of lean construction principles. The successful implementation of the system in China was analyzed and the results suggest that heat maps which use dimensional and temporal positioning data, could be an efficient and convenient method to define workspace of crews onsite [1].

Workspace is a key resource in construction sites that are required by construction activities [2]. Workspace in construction plays important roles for both space management in construction and lean approaches to be applied. Workspace generation by activities onsite requires spatiotemporal data necessary to represent in four dimensions but it is difficult for project managers to obtain spatiotemporal data manually [3] due to large efforts of data collection. There are attempts to automate generation of workspace to reduce the amount of data entered by the user but much of the research still remains in the planning phase and those space requirements with spatial information have rarely been compared with real situations of workers conducting tasks onsite.

We applied a Bluetooth Low Energy (BLE) based indoor positioning monitoring system in construction [4] and developed the method to generate the heat maps of workers’ onsite presence, aiming to outline their workspace based on density of detected points. We believe through this research, the construction industry can benefit from automated workspace detection based on heat map generation from real-time tracking systems, which complements other lean approaches in construction with goals to enhance management efficiency and eliminate the waste.

Our test from a case study in China returned promising results. The location of the case was in a Chinese city close to the capital Beijing. The construction project was undergoing a large amount of MEP work at the time of testing in a large shopping mall building. Our research suggested that even though the BLE real-time tracking system may lack accuracy in detail of a single detected point with right coordinate at the current stage, the heat maps based on density of movement point can still outline general areas of movement in real situations. Therefore, it provides evidence that heat maps can be an efficient approach to define overall workspace during any given time of the day.

For instance, the heat map with hourly intervals can easily identify which hour of the day could be the workers’ most likely-to-work time based on their movement density onsite (Figure 1).

Figure 1: Hourly heat-map changes over time for MEP worker No. 16

Furthermore, heat maps can also be conducted in detail to interpret what has happened onsite based on the workers’ coordinate information. Figure 2 demonstrates the captured workspace where one MEP worker has conducted his task between 2pm to 4pm with a high level of uninterrupted presence detected during this time. The straight line during the 2pm to 3pm time interval outlines a fuzzy coverage of the pipeline conducting area, which is correlated to this worker’s onsite uninterrupted presence [5]. It is reasonable to think that with deeper color of the heat map, the worker is more likely to conduct value-adding activities because higher frequency of position points is detected within the deeper-colored area. This also suggests that the underlying workspace during that hour was the required working space for the MEP worker in practice.

Figure 2: Heat maps in detail between 2pm to 4pm of MEP worker No. 16

Using the real-time indoor positioning system to generate heat maps of workers’ movement in construction can give us some practical use cases for future research prospects. For instance:

• Analyzing takt areas. One key objective of lean thinking is to remove variability [6]. Frandson et al [7] have proposed that an important characteristic of TTP is for a set amount of time (Takt time), each trade is committed to finish their tasks in each assigned area (Takt area). For a project that implements Takt time planning, heat maps can automatically detect the workspace over time and compare them with the assigned Takt area, to see if they match in real practice. Furthermore, heatmaps can also be generated to visualize how buffer capacity between tasks is followed and provide in-time feedback for site managers to adjust resources to ensure smooth construction flows.
• Identifying threshold values for workers’ presence in workspace. Another important objective of lean thinking is to eliminate waste onsite. Waste refers to the activities that consume resources but cannot add value for the process. Zhao et al. [5] have proposed the concept of threshold value for uninterrupted presence onsite where they reasoned that a worker needs to stay at a work location for a certain period so that it can be counted as uninterrupted time, which is correlated with value-adding time. Because heat maps are generated on a basis of the density of onsite presence for identification of workspace, the actual threshold minute should be calculated according to the heat map of that worker in question where his or her workspace is detected. Therefore, the density of the presence of that worker in question can be an indicator of the actual threshold value.
• Forecasting potential workspace congestion. Akinci et al. [2] have established an automated mechanism where they aimed to carry out project-specific spaces required by site activities. Heat maps represent the workers’ real presence onsite that includes those dimensions to outline an estimation of workspace at a specific work location and timestamp. Time-space conflicts in construction happen frequently where task activities have a temporal aspect [3]. Heat maps generated by real-time tracking systems can be used for predicting interfering degrees among different tasks during the same time because the system supports multi-resource tracking and generation of heat maps based on overall movement density of given numbers of workers.

Our findings of the heat map application indicate that it can provide both dimensional and temporal data through an automated and illustrative process in construction. Onsite temporal data such as task-level uninterrupted presence indices [8] connects with value-adding level of task performance. Onsite dimensional data such as working space area links with resource management such as labor, equipment and material. Heat maps that contain spatiotemporal data of both can generate a more straightforward space-loaded production model where value-adding activities can be potentially detected based on movement density of workers, and onsite resource handling can be optimized. The heat-map color change figure can also help site managers to quickly identify the intensive working activities taking place at a certain time and location, and a dynamic heat map figure can be used to forecast potential workspace congestion in real practice. In future research, the study could be extended to address further applications and limitations of the heat maps. For instance, if a project's production system has not been optimized, excess motion or waste in the heat maps should be positioned (e.g. where and when) to fast pinpoint the root causes. Site managers may need to compare heat map results with actual production results and problems reported on site and decide if lean interventions should be taken on accordingly.

REFERENCES

1. J. Zhao, O. Seppänen, A. Peltokorpi, Applying Heat Maps to Define Workspace in Construction Based on Real-Time Tracking System With Coordinate Positioning Information,. annual conference of the international group for lean construction. (2020).
2. B. Akinci, M. Fischer, J. Kunz, Automated generation of work spaces required by construction activities, J. Constr. Eng. Manage. 128 (2002) 306-315.
3. B. Akinci, M. Fischen, R. Levitt, R. Carlson, Formalization and automation of time-space conflict analysis, J. Comput. Civ. Eng. 16 (2002) 124-134.
4. B. Badihi, J. Zhao, S. Zhuang, O. Seppänen, R. Jäntti, Intelligent construction site: on low cost automated indoor localization using bluetooth low energy beacons, .2019 IEEE conference on wireless sensors (ICWiSe). (2019) 29-35.
5. J. Zhao, O. Seppänen, A. Peltokorpi, B. Badihi, H. Olivieri, Real-time resource tracking for analyzing value-adding time in construction, Automation in construction. 104 (2019) 52-65.10.1016/j.autcon.2019.04.003.
6. H.R. Thomas, M.J. Horman, de Souza, Ubiraci Espinelli Lemes, I. Zav ski, Reducing variability to improve performance as a lean construction principle, Journal of Construction Engineering and Management. 128 (2002) 144-154.https://doi.org/10.1061/(ASCE)0733-9364(2002)128:2(144).
7. A. Frandson, O. Seppänen, I.D. Tommelein, Comparison between location based management and takt time planning, .the 23rd annual conference of the international group for lean construction. (2015).Perth, Australia.http://iglc.net/Papers/Details/1181.
8. J. Zhao, E. Pikas, O. Seppänen, A. Peltokorpi, Using real-time indoor resource positioning to track the progress of tasks in construction sites, Frontiers in Built Environment. 7 (2021) 59.