2021.04.21
NTT COMWARE Launches SmartMainTech, New Infrastructure Maintenance Service Using Digital Technologies
~ Contributing to improvements in the sustainability of social infrastructure ~
NTT COMWARE CORPORATION (Head office: Minato-ku, Tokyo; President: Satoshi Kurishima; hereinafter NTT COMWARE) started to provide SmartMainTech®, a new business brand, on April 21, under which it will comprehensively support a series of administrative operations including the planning of new facilities and the renewal and disposal of existing facilities, and will seek to improve the sustainability and resilience of social infrastructure. This will be achieved through the application of digital twin technology to facilities and the implementation of the digital transformation (DX) of facility operations in the social infrastructure field.
NTT COMWARE has been developing and operating systems related to the management of a large volume of communication equipment owned by the NTT Group. In addition, it has been supporting diverse maintenance services of social infrastructure such as communication facilities, power generation facilities and roads through the social implementation of artificial intelligence, including the Deeptector® image recognition AI, and data science technology for the infrastructure industry. SmartMainTech, which NTT COMWARE has started to provide, is a total solution brand aimed to facilitate new digital technology-based infrastructure maintenance for the operators of social infrastructure businesses such as energy and urban development, as well as communication, by systemizing technologies utilized for these infrastructure areas and relevant expertise.In solutions provided under this brand, efforts will be made to improve facility utilization rates and optimize capital investments through the application of the digital twin technology to social infrastructure facilities, and to contribute to simultaneous improvements in both on-site operation efficiency and operation safety through DX of facility-related operations. In addition, through these efforts, facility maintenance and operation cycles will be supported totally.
In solutions provided under this brand, efforts will be made to accelerate the enhancement of product lines, namely, a group of products and services that will be offered under the solution brand, going forward, to comprehensively support the lifecycle process of facilities. To start with, NTT COMWARE launched the Infratector® typeC, an AI-based construction inspection service, and the Field Collabo, a service that contributes to productivity improvements through collaboration between frontlines and offices, today.
The Infratector typeC is Japan's first construction inspection-specific AI service dedicated to the construction of communication facilities*1. Construction operators take photos of what has been constructed according to photo shooting guide functions that respond to a range of inspection standards and shooting rules, and determine whether or not construction has been completed properly by utilizing NTT COMWARE's Deeptector image recognition AI. By applying NTT COMARE's AI technology to construction inspection operations at construction sites, operators can improve productivity and realize DX without requiring any skills.(See Appendix 1)
*1: It was confirmed that no similar products were available in the utility pole construction market (in terms of setting locations and angles, etc.) (Based on a NTT COMWARE survey, as of February 2020)
The Field Collabo is a service that realizes the provision of remote construction support and the real-time sharing and visualization of progress in construction by establishing a connection between the operation frontlines and offices in infrastructure maintenance work. It provides video-based operations support remotely and facilitates the introduction of new digital tools to frontline operations by combining business chat and operations management, while at the same time keeping on-site administrators from being troubled by complicated construction management operations. (See Appendix 2)
NTT COMWARE will seek to develop solutions and cultivate target markets by enhancing AI and data science technologies and through collaborative creation with business partners, and will help social infrastructure business operators solve the issues they need to tackle. By doing so, it will contribute to the creation of a smart world. It aims to achieve the sales target of 10 billion yen (cumulative) from SmartMainTech in five years.
Appendix 1
1. Service Overview
The Infratector typeC is Japan's first service utilizing artificial intelligence (AI) that specializes in the inspection of telecommunication facilities construction*1. Construction operators take photos of what has been constructed according to photo shooting guide functions that respond to a range of inspection standards and shooting rules, and determine whether or not construction has been completed properly by utilizing NTT COMWARE's Deeptector® image recognition AI. (The determination program is patent pending.)
Infratector typeC earned high marks from telecommunications construction companies that used it on a trial basis for improving inspection quality and for its usefulness. It has thus been confirmed to produce an effect of improving efficiency in construction inspection.
(Refer to 3. Effect of Introduction of Infratector typeC)
The Infratector typeC offers models learned by AI as Software-as-a-Service (SaaS) type services. This means that corporate users do not need to prepare teaching data or to carry out tuning*2 and that an AI service that conforms to the industry standard can be quickly introduced at a low cost. Users, namely construction workers, can use it immediately simply by installing the app on their smartphones and other terminals.
Today, improving productivity in the field of social infrastructure is seen as an issue. Especially in terms of telecommunication infrastructure, a major trend shift to 5G is apparent. The shortage of personnel for wireless facilities construction is becoming increasingly serious. This has created an urgent need to shift human resources from conventional optical fiber installations to the rapidly increasing construction of 5G networks, while providing support for engineers.
In answer to these issues facing the overall industry, the Infratector typeC automates construction processes with the use of AI technologies to accelerate digital transformation (DX). It helps improve productivity and ensures that these processes can be performed without a skill requirement.
*1: It was confirmed that no similar products were available in the utility pole construction market (in terms of setting locations and angles, etc.) (Based on a NTT COMWARE survey, as of February 2020)
*2: In the case of using learned models
As of April 21, 2021, it supports inspections at telecommunications construction sites. Support will be successively expanded.
Figure 1: Schematics before and after introduction of Infratector typeC
2. Features
The Infratector typeC consists of a photo-taking guide function and an AI function. Supporting the stable operation of telecommunication infrastructure, quality inspection of construction is important and so time-consuming as to involve human labor. The Infratector typeC has three features that resolve issues such as an engineer shortage, transfer of expertise, and standardization of inspection quality.
(1)A guide to easy photo-taking for anyone
The photo shooting guide function and the AI functions are available from a user-friendly smartphone app. The photo shooting guide function displays a photo-shooting frame optimized for taking a photo of construction on the smartphone screen to make it easy for anyone to take proper photos.
(2)High precision AI to automatically examine the conformity of construction
On the basis of the photo of construction results, the AI automatically determines whether or not they meet the standards. It employs NTT COMWARE's Deeptector image recognition AI. That performs not only detection of objects from the photo but the processing of complicated conditions for determination including colors and lengths of objects and the order of attachment. It accurately replaces complicated inspections carried out by humans.
(3)Real-time determination on conformity of construction results
On the basis of photos of construction taken with the help of the photo shooting guide function, the AI determines whether or not construction results conform to the standards on a real-time basis. Instant learning of inspection results paves the way for on-the-spot correction to errors in photography or in construction.
Figure 2: Functions of Infratector typeC
3. Effect of Introduction
The trial use by telecommunications construction companies has confirmed that the introduction of Infratector typeC produces three different effects.
(1)Photo-shooting skills no longer needed
The Infratector typeC helps reduce labor in extensive areas. For example, it reduces the costs involved in the creation and distribution of manuals after an addition or alteration to the construction completion standards as well as the costs of training and learning. Supporting changes in construction completion standards following the expansion of 5G construction, it facilitates support for engineers.
(2)Enhancing inspection efficiency
The photo shooting guide function shortens the time for taking photos and reduces errors in taking photos. The AI function improves inspection efficiency and accuracy. It covers an increased number of inspections following expansion of 5G construction.
(3)Helping achieve zero reworks
Real-time learning of inspection results on the construction site helps reduce reworking to zero.*3 Streamlining of the construction inspection process means that more construction projects can be completed. It will accelerate the shift of human resources to 5G construction, which is increasing rapidly.
*3: Zero reworks were confirmed in the trial use by a number of companies in the telecommunications construction industry.
4. Launch Date
April 21, 2021
5. Fee
Subject to quotation on a case-by-case basis according to the scale of introduction to the corporate user and the number of construction inspections
6. Service Website
https://www.nttcom.co.jp/smtech/infratectortypec/
Appendix 2
1. Service Overview
Field Collabo is a cloud service that takes advantage of communication infrastructure technologies nurtured through support for NTT's telecommunication infrastructure to link field workers conducting infrastructure maintenance work and the office, enabling remote support for the work and real-time sharing and visualization of its progress.
It offers three functions. First, the remote support function allows the field manager in charge of managing the worksite to give audiovisual support to field workers on the site. Second, the chat report function enables field workers to easily report the work process using a chat tool. And third, the work progress management function automatically saves the messages exchanged on a project-by-project basis and visualizes the work progress. The use of Field Collabo means to help introduce new digital tools to worksites and frees managers from cumbersome work management duties.
Today, as more and more social infrastructure ages, infrastructure maintenance and other related projects are increasing. Meanwhile, at work sites, readiness for new technologies is required. In addition, the regulation on the maximum overtime working hours is set to apply to the construction industry. It is requisite to streamline on-site operations with the use of digital technologies. Generally in construction work, it is mandatory to appoint a field manager, who is in charge of ensuring workers' safety. As a result of an increase in the number of work projects and in items to be reported and managed, such as work progress, evidence photos and checklists, the number of worksites and items to be managed by a single field manager is increasing. Field Collabo produces a seamless link between field workers and the office operations of field managers. It provides them with an environment for working in a single team to streamline both field operations and managerial operations.
Figure 1: Schematic of Field Collabo
In the future, Field Collabo is planned to be linked with Docomo AI Agent API®. The verification process aimed at commercialization is underway. The linkage with Docomo AI Agent API will pave the way for the creation of reports based on voice input and an interactive review on checklists based on voice reading, among other features. This will mean enhanced productivity and safety through, for example, hands-free reporting from field workers. NTT COMWARE will offer Field Collabo with a view towards facilitating communication in maintenance work and building a smart city.
2. Features
(1)Remote work support using video and voice
With the remote support function, field workers take photos of the site using smartphones and other devices and share the views with the field manager at a remote location. Then, the field manager can give advice and instructions suited to the circumstances on site, check the results and perform safety checks. In addition, messages exchanged through this function are automatically stored on the work management database in such a way that they are associated with relevant projects. This allows these messages to be used as evidence of work.
(2)Reporting by business chat
After a work project is registered, a chat room for communication between the field workers and the field manager is automatically created on the business chat they are using. They can send reports with the use of the same tool as the one they usually use for business chats. This means that they can deliver reports readily and easy without using any other app.
(3)Work progress management
The log of remote work support using video and voice and chat reports are automatically stored on the work management database. They are incorporated into the progress of work. That allows the field manager to track the progress of multiple work projects running in parallel and identify any problems. The field manager will be able to give timely and appropriate instructions to field workers. In addition, they can be used as evidence of work.
3. Service Structure
Field workers use a business chat app on their mobile terminals to use the service. The API linkage allows Field Collabo to be linked with various field support tools.
4. Concept of use and effect of introduction
When a field worker arrives at the site, they report the start of work by chat. If any question or point to be checked occurs during the work, they use the remote support function to share video and sound with the field manager for checks and discussions. After work is complete, they send a photo attachment by chat. The messages, photos and others exchanged are imported one by one to the work progress database.
Figure 2: Circumstances of use of Field Collabo
Field Collabo was used on a trial basis by telecommunications construction companies. That has demonstrated that it will bring to field managers higher efficiency in the form of smoother work progress management and higher work quality, such as avoiding omitted processes. For field workers as well, the service is confirmed to increase efficiency through chat reports and to heighten the frequency of reports, communications and consultations as well as the quality level by facilitating communications with field managers. It is expected to reduce labor for field managers by around 20% to 30% and that for field workers by approximately 10% to 20%.
5. Launch Date
April 21, 2021
6. Fee
Subject to quotation on a case-by-case basis according to the scale of introduction to the corporate user and the conditions of use
7. Service Website
https://www.nttcom.co.jp/smtech/fieldcollabo/
*SmartMainTech is a registered trademark of NTT COMWARE CORPORATION.
*Deeptector is a registered trademark of NTT COMWARE CORPORATION.
*Infratector is a registered trademark of NTT COMWARE CORPORATION.
*Field Collabo is a pending trademark of NTT COMWARE CORPORATION.
*Docomo AI Agent API is a registered trademark of NTT DOCOMO, INC.
*Other company names, product names and service names listed herein are the trademarks or registered trademarks of their respective companies.
