Completed Proof-of-Concept to Streamline Confirmation of Metropolitan Expressway Closure Plans
TransRecog completed a PoC for a tool that automates the verification of road closure plans for the Metropolitan Expressway. It reads Excel data to detect rule violations, significantly reducing manual effort and errors.
📋 Article Processing Timeline
- 📰 Published: April 14, 2026 at 19:00
- 🔍 Collected: April 14, 2026 at 10:31
- 🤖 AI Analyzed: April 19, 2026 at 18:48 (128h 16m after Collected)
TransRecog Inc. (Headquarters: Minato-ku, Tokyo, Representative Director & CEO: Takaaki Kobayashi), as part of the 'Shutoko Open Innovation Challenge 2025' hosted by Metropolitan Expressway Co., Ltd., developed a confirmation support tool aimed at streamlining road closure plan coordination tasks, and confirmed its effectiveness in a proof-of-concept experiment conducted in FY2025.
In this demonstration, by directly reading the road closure plan sheets used internally by Shutoko and cross-referencing them with baseline data organizing location information and closure rules, multiple confirmation tasks previously performed manually by staff were automated. This confirmed the tool's effectiveness in reducing verification man-hours and preventing oversights.
Background
In the task of coordinating road closure plans, it is necessary to check for contradictions or interferences based on preferred dates, target sections, reserve days, and connection relationships submitted by multiple departments. The subjects of confirmation include many perspectives, such as combinations that cannot be implemented simultaneously, impacts of JCT (junction) closures, directional differences like inner/outer loops or up/down lines, time interpretations crossing midnight, and inconsistent notations of location or section names.
When these checks are performed solely by hand, not only are the confirmation man-hours large, but they also tend to rely heavily on the experience and attention of the staff, increasing the risk of oversights and missed checks, especially during busy periods. Furthermore, in the coordination field, it is important not only to judge that 'there is a problem' but also to be able to explain 'why it is a problem' and 'which rule it is based on'.
Overview of the Proof-of-Concept Experiment
In this demonstration, we built a system that directly reads the Excel format plan sheets operated internally at Shutoko and automatically performs checks such as the following while cross-referencing with relevant master data:
- Confirmation of combinations of road closures that must not be implemented simultaneously
- Confirmation of connection relationships affected during JCT closures
- Confirmation of duplicates or time zone discrepancies within the same day or month regarding the same entrance/exit
- Extraction of unmatched locations due to inconsistent notations or non-registration in the master
- Output of an Excel report organizing violations, warnings, mismatches, and parsing warnings
The output results consist of a summary sheet to grasp the overall numbers and a detailed sheet organizing violations, warnings, and mismatches by purpose, forming a structure that makes it easy for the staff to return to the original form to check.
Results Confirmed in the Proof-of-Concept Experiment
Through this demonstration experiment, we built a confirmation platform that can perform everything from reading plan sheets to judging, organizing results, and outputting reports in a batch. This reduces the burden of simple verification tasks on the staff, making it easier for them to focus on making decisions and explaining to stakeholders.
We also confirmed that check points easily missed by hand, such as direction dependency, day-crossing, inconsistent notations, and connection relationships, can be stably extracted based on rules.
Furthermore, because the revised plan sheets can be rechecked repeatedly under the same conditions, we confirmed its effectiveness as a confirmation platform applicable to both annual and monthly plan coordination. In addition, organizing the judgment results as an Excel report made it easier to explain 'from what perspective there is a problem and what needs to be confirmed,' contributing to an improvement in explainability on site.
Future Developments
Based on the results confirmed in this demonstration, TransRecog will independently proceed with studies toward functional expansion and practical application as its own solution in the next stage.
First, to make it easier for site personnel to try it out, we plan to proceed with the implementation of a UI (operation screen) to replace command execution. At the same time, we will expand the scope of application in practical work by expanding target bureaus and routes, responding to differences in forms, and adding batch execution and history comparison functions.
Furthermore, in the future, based on this check support, we aim to develop it into coordination support through AI optimization that considers priorities and constraints, as well as presenting candidate dates and time slots that resolve conflicts among multiple road closure requests. Through this, we will evolve it from a mere confirmation support tool into a platform that supports overall plan coordination.
Comments
Takaaki Kobayashi, Representative Director & CEO, TransRecog Inc.
'As we proceeded with the demonstration, I realized once again that the coordination of road closure plans includes far more confirmation points than imagined, placing a heavy burden on the staff. I am glad that, with the cooperation of Metropolitan Expressway Co., Ltd., we were able to confirm the effectiveness of confirmation support using the site's forms directly as input. We will first prepare a form that is easy to try on site, gradually expand the targets, and aim to realize coordination support utilizing AI optimization in the future.'
Company Overview
Company Name: TransRecog Inc.
Representative: Takaaki Kobayashi, Representative Director & CEO
Location: 5-23-1 Shimbashi, Minato-ku, Tokyo
Business Content: Business support software
In this demonstration, by directly reading the road closure plan sheets used internally by Shutoko and cross-referencing them with baseline data organizing location information and closure rules, multiple confirmation tasks previously performed manually by staff were automated. This confirmed the tool's effectiveness in reducing verification man-hours and preventing oversights.
Background
In the task of coordinating road closure plans, it is necessary to check for contradictions or interferences based on preferred dates, target sections, reserve days, and connection relationships submitted by multiple departments. The subjects of confirmation include many perspectives, such as combinations that cannot be implemented simultaneously, impacts of JCT (junction) closures, directional differences like inner/outer loops or up/down lines, time interpretations crossing midnight, and inconsistent notations of location or section names.
When these checks are performed solely by hand, not only are the confirmation man-hours large, but they also tend to rely heavily on the experience and attention of the staff, increasing the risk of oversights and missed checks, especially during busy periods. Furthermore, in the coordination field, it is important not only to judge that 'there is a problem' but also to be able to explain 'why it is a problem' and 'which rule it is based on'.
Overview of the Proof-of-Concept Experiment
In this demonstration, we built a system that directly reads the Excel format plan sheets operated internally at Shutoko and automatically performs checks such as the following while cross-referencing with relevant master data:
- Confirmation of combinations of road closures that must not be implemented simultaneously
- Confirmation of connection relationships affected during JCT closures
- Confirmation of duplicates or time zone discrepancies within the same day or month regarding the same entrance/exit
- Extraction of unmatched locations due to inconsistent notations or non-registration in the master
- Output of an Excel report organizing violations, warnings, mismatches, and parsing warnings
The output results consist of a summary sheet to grasp the overall numbers and a detailed sheet organizing violations, warnings, and mismatches by purpose, forming a structure that makes it easy for the staff to return to the original form to check.
Results Confirmed in the Proof-of-Concept Experiment
Through this demonstration experiment, we built a confirmation platform that can perform everything from reading plan sheets to judging, organizing results, and outputting reports in a batch. This reduces the burden of simple verification tasks on the staff, making it easier for them to focus on making decisions and explaining to stakeholders.
We also confirmed that check points easily missed by hand, such as direction dependency, day-crossing, inconsistent notations, and connection relationships, can be stably extracted based on rules.
Furthermore, because the revised plan sheets can be rechecked repeatedly under the same conditions, we confirmed its effectiveness as a confirmation platform applicable to both annual and monthly plan coordination. In addition, organizing the judgment results as an Excel report made it easier to explain 'from what perspective there is a problem and what needs to be confirmed,' contributing to an improvement in explainability on site.
Future Developments
Based on the results confirmed in this demonstration, TransRecog will independently proceed with studies toward functional expansion and practical application as its own solution in the next stage.
First, to make it easier for site personnel to try it out, we plan to proceed with the implementation of a UI (operation screen) to replace command execution. At the same time, we will expand the scope of application in practical work by expanding target bureaus and routes, responding to differences in forms, and adding batch execution and history comparison functions.
Furthermore, in the future, based on this check support, we aim to develop it into coordination support through AI optimization that considers priorities and constraints, as well as presenting candidate dates and time slots that resolve conflicts among multiple road closure requests. Through this, we will evolve it from a mere confirmation support tool into a platform that supports overall plan coordination.
Comments
Takaaki Kobayashi, Representative Director & CEO, TransRecog Inc.
'As we proceeded with the demonstration, I realized once again that the coordination of road closure plans includes far more confirmation points than imagined, placing a heavy burden on the staff. I am glad that, with the cooperation of Metropolitan Expressway Co., Ltd., we were able to confirm the effectiveness of confirmation support using the site's forms directly as input. We will first prepare a form that is easy to try on site, gradually expand the targets, and aim to realize coordination support utilizing AI optimization in the future.'
Company Overview
Company Name: TransRecog Inc.
Representative: Takaaki Kobayashi, Representative Director & CEO
Location: 5-23-1 Shimbashi, Minato-ku, Tokyo
Business Content: Business support software