What is data utilization in the construction industry? A clear explanation of use cases and implementation methods.

The background behind the need for data utilization in the construction industry

The use of data in the construction industry is not simply a matter of it being "convenient," but is becoming an indispensable practice for on-site operations, driven by the industry's structure and the need to adapt to regulations. In particular, in recent years, challenges that are difficult to address with traditional methods have become apparent, such as the difficulty of skill transfer due to labor shortages, inefficiencies caused by dispersed information management at each site, and the need for more sophisticated working time management due to work style reforms.

To address these challenges, it is crucial to create a system that accumulates and integrates data and utilizes it for decision-making at both the operational and management levels. Here, we will outline three main reasons why data utilization is becoming increasingly important in the construction industry.

Labor shortages and challenges of skills transfer

The construction industry is facing both an aging workforce and a shortage of young workers, making it increasingly difficult to maintain on-site productivity simply by "increasing the number of people." As a result, the fewer people involved, the more directly planning errors and rework translate into increased costs.

The key here is to transform the intuition and experience of veterans into reproducible information. For example, if you record data such as the standard man-hours for each type of work, the extent of delays due to weather and delivery conditions, and signs of defects visible from photos and inspection results, even younger employees can improve the quality of their decision-making.
Skill transfer is not only about education, but also directly impacts the accuracy of estimates and plans. If tacit knowledge remains confined to individuals, each time the person in charge changes, they have to relearn everything from scratch. However, if it is digitized and incorporated into standard procedures and checklists, it can be built up as an organization.

Data silos and duplicate data entry

At construction sites, drawings, schedules, daily reports, photos, progress reports, cost estimates, and safety documents tend to be stored in separate locations. When paper, Excel spreadsheets, email attachments, and individual systems are all mixed together, time is wasted just searching, checking, and transcribing.
The problem with siloed systems is not just inconvenience, but also the fact that errors due to not knowing the latest version, or transcription errors, can have repercussions on quality, cost, and safety. For example, if process changes are not reflected in costs or staffing, the factory floor is forced to make drastic recovery efforts once they realize the problem.

The value of integration and centralized management lies not just in data analysis, but in eliminating duplicate data entry and ensuring everyone sees the same facts. Simply standardizing where information is stored and how it's linked can significantly reduce indirect effort and rework.

▼I want to know more about siloing
⇒ Siloization | Glossary

Work style reform and advanced working time management

Even with progress being made to comply with the 2024 overtime work limit regulations, the construction industry continues to face the challenge of optimizing working hours while managing sites with limited personnel. The key is to understand where time is actually being wasted and to improve the accuracy of planning and the speed of corrective actions in operation.
In many cases, the cause of overtime cannot be identified by looking only at attendance records. Overtime is often caused by a chain reaction of factors such as process delays, insufficient preparation, delays in material procurement, and re-inspections. By looking at processes, staffing, output, defects, and delivery records in conjunction, it becomes easier to detect early warning signs before delays become apparent.

The purpose of using data is not for monitoring, but to quickly identify and correct structural issues that lead to unreasonable work practices. To gain the acceptance of those on the ground, it is essential to implement the system in a way that allows them to see the benefits, such as correcting workload imbalances and improving work processes.

Back to Top

What can be achieved by utilizing data in the construction industry | Use cases

In the construction industry, data utilization goes beyond mere information storage; it's being used in various aspects of on-site operations, including process management, safety management, quality control, and equipment management. The benefits of data utilization tend to be particularly evident in areas where daily decisions are frequent and early detection of anomalies can minimize losses.
Here, we introduce four typical use cases for data utilization in the construction industry.

Optimization of construction plans and visualization of progress and costs

The best textbook for construction planning is the track record of "similar past projects." By accumulating data such as actual man-hours for each type of work, outsourcing ratios, delivery conditions, and delays due to weather, and comparing this with the project conditions, the accuracy of construction period, cost, and personnel plans can be improved.
When creating dashboards for progress, output, and costs, the goal is not simply to display numbers, but to track the differences from the plan and identify where those differences are coming from. For example, being able to view data by work section, work type, or subcontractor allows for more concrete action to be taken.
Furthermore, visualizing progress, man-hours, and staffing levels together makes it easier to identify imbalances in workload and signs of increased overtime.
External data is also useful. Incorporating weather, material prices, and traffic restrictions increases the explainability of delays and cost increases, and allows for earlier coordination with clients and subcontractors. The key is to connect the analysis results to operations such as the next weekly meeting and personnel changes.

Safety management (hazard prediction and near-miss analysis)

Safety is an area where "reflection after an incident" is common, but data collection clearly identifies key countermeasures. By compiling near-miss and accident reports by type of work, time of day, weather, location, and cause classification, it becomes possible to narrow down the focus of patrols and the content of training.
Furthermore, using data such as camera footage, entry/exit records, and location information, it's possible to detect signs of dangerous behavior and issue warnings. However, the key is designing a system that is acceptable to those on the ground. If it gives the impression of being overly strict, reports will decrease, and the quality of the data will decline.

From an operational standpoint, it's more practical to avoid issuing too many alerts and focus on high-risk cases. Prioritizing cases that are likely to lead to serious accidents and improving rules and thresholds in collaboration with on-site personnel will make the system easier to implement.

Improving the efficiency of quality control and inspection (photos and forms)

While evidence is crucial for quality control, organizing photos and creating reports often becomes a significant burden on on-site staff. By centralizing photos in the cloud and tagging them by work area, work type, location, and date, the time spent searching for photos and submitting them can be drastically reduced.
Standardizing the input format for inspection results makes it possible to see the causes of quality variations. For example, being able to track where non-conformities occur, the collaborating companies involved, and the process stage allows corrective actions to shift from being "one-off" to "preventive."

Remote site inspections, drones, and point cloud data not only reduce the time required for verification but also lower the risk of oversights. From a data utilization perspective, the key is to link drawings and BIM/CIM elements with photographs and inspection records, so that anyone can find the same evidence.

Operational analysis and preventive maintenance of construction machinery

For construction equipment, the operating rate directly impacts profits. By collecting data on operation/stoppage, idling, fuel consumption, and breakdown logs, unnecessary waiting times and inefficient operations become visible, making it easier to reduce both fuel and rental costs.

The value of preventive maintenance lies not in the failure itself, but in reducing the losses caused by "stoppages." On-site planning can be instantly disrupted by construction equipment downtime, leading to delays, overtime, and additional costs. By detecting signs of abnormalities and transitioning to planned maintenance, downtime can be minimized.
When implementing the system, it's more practical to start with machines that have long operating hours or those that are most affected by malfunctions, rather than immediately implementing it across all models and all locations. If the analysis results can be incorporated into the vehicle dispatch plan and inspection cycle revisions, it becomes easier to explain the return on investment.

Back to Top

Steps to promote data utilization in the construction industry | DX steps

When it comes to utilizing data in the construction industry, it's crucial to proceed gradually rather than trying to implement everything at once. Trying to create a large-scale system from the beginning will increase the workload on the ground and make it difficult to establish and maintain its operation.
The basic process involves first digitizing, collecting, and storing on-site information, then connecting and integrating various business data, and finally progressing to visualization, analysis, and automation. Here, we will outline the typical steps involved in advancing data utilization in the construction industry.

Data collection and storage (digitalization of the field)

The first step is to replace paper, verbal, and individual management with digital input, creating a system where data is permanently recorded. Specific examples include smartphone-based daily reports, cloud storage of photos, sharing of drawings, and digitizing checklists.
At this stage, ease of input is paramount. If input is cumbersome, omissions and delays will increase, ultimately resulting in unusable data. Reduce the burden through operational design, such as increasing the number of selection options and reducing the number of inputs, to match the workflow on site.
In addition, we will improve tag design and data quality. By standardizing essential fields such as project ID, work area, work type, date, and subcontractors, and minimizing inconsistencies in terminology, the accuracy of subsequent data aggregation will be greatly improved. Collecting data neatly from the start will reduce the cost of subsequent analysis.

Data integration (common ID, master data, and integration platform)

If the collected data is scattered across different departments and tools, it will ultimately require re-transferring the data. Therefore, we define common IDs and master data such as project ID, work area, work type, and subcontractor, and connect process, cost, quality, safety, and labor across all systems.
The goal of the integration is not just to make the entire company more transparent. At the operational level, the value lies in creating a system where information is automatically aligned, such as linking process changes to staffing and costs. This reduces the time spent "checking which is correct."
Integration will utilize mechanisms such as APIs and ETL to the extent possible, eliminating manual data transfer. Initially, instead of aiming for full system integration, eliminating the most impactful instances of duplicate data entry will lead to a more favorable return on investment.

In the construction industry in particular, systems tend to be separated by purpose, such as site management, cost management, attendance management, and accounting management, making integration design complex. Using iPaaS is an effective way to streamline such complex system integrations. For example, using an iPaaS like HULFT Square makes it easier to manage API and file transfer in a unified manner. By standardizing data integration platform without increasing individual development, it's possible to achieve a configuration that is easy to expand while keeping maintenance burdens low.

▼I want to know more about the API
⇒ API｜Glossary

Designing a data infrastructure to support its utilization (BIM/CIM, process, cost, photos, IoT)

In the foundational design phase, structured data such as processes, costs, attendance records, and inspection results, along with unstructured data such as drawings, photographs, videos, and point clouds, are handled within the same "project context." Storing them separately would significantly increase the cost of relinking them later.

When using BIM/CIM as the core, linking model elements and location information with photos, completed work, inspection records, and IoT (location/operation) using site IDs improves searchability and explanatory power. For example, it becomes possible to quickly retrieve "construction records and inspection results for this part."
Furthermore, access control, version control, and audit logs are crucial from an operational standpoint. A breakdown in the management of the latest version of drawings can lead to quality issues, and the inability to track who viewed what and when makes it difficult to explain problems when they occur. Designing not only the technology but also the operational rules together creates a truly "usable platform."

Analysis, prediction, and automation (using AI)

Once we have the ability to handle the necessary data across different systems, the first step is to stabilize our understanding of the current situation through visualization. We narrow down the key KPIs to a few, such as process delays, cost overruns, increased overtime, and safety risks, and incorporate them into daily and weekly decision-making. We don't just stop at viewing the data; we also design rules for responding when discrepancies occur.

The next step involves detecting early signs of delays, overruns, and accidents, and recommending improvement plans. AI is not magic; it is most effective in areas where reproducible patterns are visible in the field. It is realistic to start with rule-based alerts and then develop them into AI while refining accuracy and operation.
Automation, such as automatic form generation and data entry assistance, is a well-suited measure because it reduces the burden on on-site staff while improving data quality. Establishing a system for continuously improving models and rules during operation ensures that the results are not merely temporary.

Back to Top

Challenges and solutions during implementation: on-site adoption, personnel, cost, and security

Data utilization is an area where stumbling blocks often lie more in on-site operations, organizational structures, and investment decisions than in technical challenges. This article summarizes typical obstacles and practical solutions.

The biggest challenge is getting it adopted by the field staff. If they feel that the input process is increasing, that they are being monitored, or that the benefits are not visible, they will not collect data. The solution is to minimize the number of input fields and create a feedback loop where the collected data is used to improve procedures and reduce paperwork at the field level. Starting with measures that reduce the burden, such as organizing photos or compiling daily reports, will make it easier for the system to be accepted.

Next is the hurdle of human resources. It's not realistic to immediately assemble a team of specialists; instead, it's more practical to divide the roles between a facilitator who understands the on-site operations and someone who can design the data, and run things on a small scale. Even when using external support, it's important to have a policy of accumulating internal knowledge, such as master data design and KPI definition, rather than simply handing everything over.

Cost and return on investment become difficult to assess when considering company-wide implementation. First, narrow down the target sites and issues, and demonstrate the effects using easily measurable metrics such as reducing duplicate data entry, indirect labor costs, rework, and overtime, thereby creating a basis for expanding the system to other areas.

Security should never be a last resort. Since we handle drawings and customer information, we incorporate access control, device management, logging, and data exfiltration prevention from the very beginning. To ensure both convenience for on-site staff and security, we clearly define access scopes for each role, designing the system so that only those who need access can see what they need.

Back to Top

Key points for successful data utilization in the construction industry

Successful companies share common characteristics: "clearly defining their objectives," "starting small and expanding horizontally," and "designing operations that connect the front lines with management." We'll cover the key points for consistently achieving results.

The objective is not defined as "collecting data," but rather as "improving what kind of decision-making." For example, by focusing on actionable objectives such as early correction of delays, nipping cost overruns in the bud, identifying the causes of overtime, or prioritizing safety risk countermeasures, you can stay focused.
A small-scale start is not just about reducing the scale, but also about narrowing the scope and "operating it to the end." Don't stop at the proof of concept (PoC); integrate it into existing operations such as weekly meetings and patrols, and decide who will oversee and who will take action. If the operation is running smoothly, the training costs will also decrease when expanding it to other areas.
To connect the field with management, it's necessary to bridge the gap between field KPIs and management KPIs. For example, the man-hours and rework reported in daily reports ultimately impact profit margins and project timelines. Showing different values for the same data—reducing the burden on the field and improving profits for management—makes it easier to gain investment and cooperation.

Finally, it's important to create a culture that adheres to data definitions. If project IDs or work type classifications are inconsistent, the analysis becomes ineffective. By adhering to the minimum rules initially established and creating a system for improvement based on feedback from the field, data utilization will become an established part of the business infrastructure rather than just a one-off measure.

Back to Top

summary

The use of data in the construction industry is a crucial initiative to improve on-site productivity, safety, quality, and labor practices, driven by labor shortages and the need to address work-style reforms. The goal is not simply IT implementation, but rather to reduce rework, waste, accidents, and overtime based on actual events occurring on-site, thereby achieving both profitability and quality.

A realistic approach would be to start with data collection through on-site digitalization, integrate it using a common ID and collaboration platform, and then gradually develop it into visualization, analysis, and prediction. Starting small and gradually expanding it while integrating it into operations will lead to the creation of sustainable on-site operations that can be run by a small number of people. The first step in construction DX is to start by "preserving" the data generated on-site in a connected format.

Back to Top

The person who wrote the article

Affiliation: Marketing Department

Yoko Tsushima

After joining Appresso (now Saison Technology), he worked as a technical sales representative, in charge of technical sales, training, and technical events. After leaving the company to return to his hometown, he rejoined the company in April 2023 under the remote work system. After gaining experience in the product planning department, he is currently in charge of creating digital content in the marketing department.
(Affiliations are as of the time of publication)