Work Pattern Analysis with and without Site-specific Information in a Manufacturing Line

Kurata Takeshi
Kurata TakeshiDeputy Director, HARC, AIST um AIST
Work Pattern Analysis
w/ & w/o Site-specific Information
in a Manufacturing Line
Takeshi Kurata1, Rei Watanabe2
Satoki Ogiso1, Ikue Mori1, Takahiro Miura1, Karimu Kato1
Yasunori Haga3, Shintaro Hatakeyama3, Atsushi Kimura3 and Katsuko Nakahira2
1 Human Augmentation Research Center, AIST, Japan
2 Faculty of Engineering, Nagaoka University of Technology, Japan
3 DENSO CORPORATION, Japan
APMS 2023
HPM and Geospatial Intelligence (GSI)
2
Kurata, T., Geospatial Intelligence for Health and Productivity Management in Japanese Restaurants and
Other Industries, APMS, pp. 206–214 (2021) doi: 10.1007/978-3-030-85906-0_23
Geospatial Intelligence (GSI) with 6M data
3
IE: Industrial Engineering, OR: Operations Research,
IoT: Internet of Things, IoH: Internet of Humans, UI: User Interface
XR: VR, AR, MR, etc. (AR: Augmented Reality, VR: Virtual Reality, MR: Mixed Reality)
RM: Raw Material, WIP: Work In Progress, SFG: Semi-Finished Goods, FG: Finished Goods
Conceptual diagram of GSI 6M information in Service and
Manufacturing sites
Study focus and research question
5
DENSO CORPORATION: Promoting activities aimed at improving both
productivity and QoW, especially the working environment, such as the thermal
environment.
DENSO & AIST: Conducting multifaceted analysis of productivity and QoW using
various acquired data while utilizing indoor GSI technology.
Focus of this study (today’s presentation): Work analysis using flow line data for
capturing a comprehensive view of the actual work behavior of each worker on the
manufacturing line.
Research question: Examine whether it is possible to start the analysis even before
site-specific information is available.
Room temperature heat map and flow line (exposed temperature and activity of each worker)
Work pattern analysis w/ GSI:
Start with or without site-specific information?
4
Work site in this study
6
Manufacturing line in a DENSO factory
• Manufacturing process of automotive work-in-progress
• Main area for analysis: manufacturing line (1,400 m2)
Measurement period: 5 days
Number of workers: 10
• Day shift: 5 (Leader, Deputy leader, Receiving, Visual inspection, Internal inspection)
• Night shift: 5 (Leader, Deputy leader, Receiving, Visual inspection, Assembly)
Manufacturing line
Break
room
Air shower
Staying plots
Manufacturing line
Air shower Break room
Indoor Positioning
7
Indoor positioning system in this study
(PDR & BLE & Map)
CE50: about 3m
Indoor positioning tech map
[16] S.Ogiso, et al., “Integration of BLE-based proximity detection with particle filter for day-long stability
in workplaces,” IEEE/ION PLANS 2023
BLE beacons
• 48 locations in the manufacturing line
• 8 locations in the break room and others
Overview of our work analysis method
8
Staying
plots
Extract “staying plots” from
flow lines
Generate the “work area transition model” by
clustering of staying plots of all shifts & workers
Obtain “work area transition instances” by
registering features of each shift & worker in
the work area transition model
Find “work patterns” by
clustering of work area transition instances
Extract exceptions from work area transition
instances by comparing to work patterns
Work area
transition model
Clustering
Work pattern A
(Cluster A) Work pattern B
(Cluster B)
Work pattern C
(Cluster C)
Clustering
Indoor
positioning
Work area transition
instances
Exception 1
(non-routine work 1)
Exception 2
(non-routine work 2)
How to generate
the work area transition model
9
K-means++
Work area transition model generated in this study
10
525 dimensions in total
• 84-dimensional staying rate features
• 441-dimensional moving rate features
How to obtain a work area transition instance w/
the work area transition model
11
Registering staying rate features and moving rate features of each shift & worker
in the work area transition model
Staying rate
features
Moving rate
features
Extract “staying plots” from
flow lines
Generate the “work area transition model” by
clustering of staying plots of all shifts & workers
Obtain “work area transition instances” by
registering features of each shift & worker in
the work area transition model
Find “work patterns” by
clustering of work area transition instances
Extract exceptions from work area transition
instances by comparing to work patterns
Find work patterns
12
Staying
plots
Exception 1
(non-routine work 1)
Work area
transition model
Clustering
Work pattern A
(Cluster A) Work pattern B
(Cluster B)
Work pattern C
(Cluster C)
Clustering
Indoor
positioning
Work area transition
instances
Exception 2
(non-routine work 2)
K-means++
Examples of work area transition instances and clustering results
13
Example of work area transition instances in the typical work patterns for each role
Assumed deployment information for each role
provided by the site manager after clustering
Clustering results of work area transition instances
and their relationship to roles
• ID 1-8: Work pattern IDs (cluster IDs)
• ID 0: Four work patterns with one instance
• 46 work area transition instances into 12 clusters
Note: Role and shift info are used only as ground
truth (not for clustering).
1 2 3
5 7 8
14
How to extract exceptions (non-routine works)
from work area transition instances
RSS: Residual Sum of Squares
Extracted exceptions of work area transition instances
15
Seven extracted exceptions of work area
transition instances along with their
exception indicators and the threshold.
Exceptions A and B for visual inspection.
5
Conclusions
16
12 work patterns and 7 non-routine works found w/o
site-specific information containing each worker's
role, shift and typical work areas of each worker
More findings by interviewing on-site
managers w/ the analysis results
(Not discussed today...)
Future works
Verification of
the applicability
in other sites
Pre-evaluation of site
improvement/design ideas by
simulation based on the work
area transition model and
work patterns
Multifaceted analysis of productivity
and QoW w/ physical-work data,
operational data, environmental
exposure data, vital sign data, and
subjective data
Work pattern analysis only w/ flow lines
based on indoor GSI (Geospatial Intelligence)
On-going other case: Express-way service area with a two-
story building
17
26 clusters of staying plots (48 workers, 20 days) Work area transition model
Examples of work area transition instances
1F
2F
1F
2F
Evaluating the work environment and physical load of factory workers
18
[15] Nakae, S., el al., Geospatial intelligence system for
evaluating the work environment and physical load of
factory workers, 45th Annual International Conference
of the IEEE Engineering in Medicine & Biology Society
(EMBC), 5 pages, 2023
Multifaceted analysis of productivity and
QoW w/ physical-work data, operational
data, environmental exposure data, vital
sign data, and subjective data
Previous related work: Simulation with a work process model
generated from flow lines and picking data in a warehouse
19
Single picking Zone picking
Myokan, T., et al., Pre-evaluation of Kaizen plan
considering efficiency and employee satisfaction by
simulation using data assimilation -Toward
constructing Kaizen support framework-, Proc.
International Conference of Serviceology (ICServ2016),
7 pages (2016)
Conclusions
20
12 work patterns and 7 non-routine works found w/o
site-specific information containing each worker's
role, shift and typical work areas of each worker
More findings by interviewing on-site
managers w/ the analysis results
(Not discussed today...)
Future works
Verification of
the applicability
in other sites
Pre-evaluation of site
improvement/design ideas by
simulation based on the work
area transition model and
work patterns
Multifaceted analysis of productivity
and QoW w/ physical-work data,
operational data, environmental
exposure data, vital sign data, and
subjective data
Work analysis only w/ flow lines
based on indoor GSI (Geospatial Intelligence)
1 von 20

Recomendados

Automated Sys. Design.11-For LinkedIn von
Automated Sys. Design.11-For LinkedInAutomated Sys. Design.11-For LinkedIn
Automated Sys. Design.11-For LinkedInHARDIK MODI
133 views60 Folien
Analyzing Operations on a Manufacturing Line using Geospatial Intelligence T... von
Analyzing Operations on a Manufacturing Line using Geospatial Intelligence T...Analyzing Operations on a Manufacturing Line using Geospatial Intelligence T...
Analyzing Operations on a Manufacturing Line using Geospatial Intelligence T...Kurata Takeshi
41 views20 Folien
Artificial Neural Network Based Graphical User Interface for Estimation of Fa... von
Artificial Neural Network Based Graphical User Interface for Estimation of Fa...Artificial Neural Network Based Graphical User Interface for Estimation of Fa...
Artificial Neural Network Based Graphical User Interface for Estimation of Fa...ijsrd.com
140 views4 Folien
Artificial Neural Network Based Graphical User Interface for Estimation of Fa... von
Artificial Neural Network Based Graphical User Interface for Estimation of Fa...Artificial Neural Network Based Graphical User Interface for Estimation of Fa...
Artificial Neural Network Based Graphical User Interface for Estimation of Fa...ijsrd.com
498 views4 Folien
The Performance Analysis of a Fettling Shop Using Simulation von
The Performance Analysis of a Fettling Shop Using SimulationThe Performance Analysis of a Fettling Shop Using Simulation
The Performance Analysis of a Fettling Shop Using SimulationIOSR Journals
517 views5 Folien
12 mf3im15 von
12 mf3im1512 mf3im15
12 mf3im15Sumit Kumar
98 views27 Folien

Más contenido relacionado

Similar a Work Pattern Analysis with and without Site-specific Information in a Manufacturing Line

Micro Planning And CNC ProgrammingFor Cylindrical Part In AMPPS von
Micro Planning And CNC ProgrammingFor Cylindrical Part In AMPPS Micro Planning And CNC ProgrammingFor Cylindrical Part In AMPPS
Micro Planning And CNC ProgrammingFor Cylindrical Part In AMPPS meijjournal
5 views6 Folien
Mechanical Engineering: An International Journal (MEIJ) von
Mechanical Engineering: An International Journal (MEIJ)Mechanical Engineering: An International Journal (MEIJ)
Mechanical Engineering: An International Journal (MEIJ)meijjournal
10 views6 Folien
Micro Planning And CNC ProgrammingFor Cylindrical Part In AMPPS von
Micro Planning And CNC ProgrammingFor Cylindrical Part In AMPPS Micro Planning And CNC ProgrammingFor Cylindrical Part In AMPPS
Micro Planning And CNC ProgrammingFor Cylindrical Part In AMPPS meijjournal
45 views6 Folien
C045061425 von
C045061425C045061425
C045061425IJERA Editor
193 views12 Folien
Optimization of Assembly Line and Plant Layout in a Mass Production Industry... von
	Optimization of Assembly Line and Plant Layout in a Mass Production Industry...	Optimization of Assembly Line and Plant Layout in a Mass Production Industry...
Optimization of Assembly Line and Plant Layout in a Mass Production Industry...inventionjournals
706 views4 Folien
Automated functional size measurement for three tier object relational mappin... von
Automated functional size measurement for three tier object relational mappin...Automated functional size measurement for three tier object relational mappin...
Automated functional size measurement for three tier object relational mappin...IWSM Mensura
450 views20 Folien

Similar a Work Pattern Analysis with and without Site-specific Information in a Manufacturing Line(20)

Micro Planning And CNC ProgrammingFor Cylindrical Part In AMPPS von meijjournal
Micro Planning And CNC ProgrammingFor Cylindrical Part In AMPPS Micro Planning And CNC ProgrammingFor Cylindrical Part In AMPPS
Micro Planning And CNC ProgrammingFor Cylindrical Part In AMPPS
meijjournal5 views
Mechanical Engineering: An International Journal (MEIJ) von meijjournal
Mechanical Engineering: An International Journal (MEIJ)Mechanical Engineering: An International Journal (MEIJ)
Mechanical Engineering: An International Journal (MEIJ)
meijjournal10 views
Micro Planning And CNC ProgrammingFor Cylindrical Part In AMPPS von meijjournal
Micro Planning And CNC ProgrammingFor Cylindrical Part In AMPPS Micro Planning And CNC ProgrammingFor Cylindrical Part In AMPPS
Micro Planning And CNC ProgrammingFor Cylindrical Part In AMPPS
meijjournal45 views
Optimization of Assembly Line and Plant Layout in a Mass Production Industry... von inventionjournals
	Optimization of Assembly Line and Plant Layout in a Mass Production Industry...	Optimization of Assembly Line and Plant Layout in a Mass Production Industry...
Optimization of Assembly Line and Plant Layout in a Mass Production Industry...
inventionjournals706 views
Automated functional size measurement for three tier object relational mappin... von IWSM Mensura
Automated functional size measurement for three tier object relational mappin...Automated functional size measurement for three tier object relational mappin...
Automated functional size measurement for three tier object relational mappin...
IWSM Mensura450 views
Application of ANFIS in Civil Engineering- A Critical Review von IRJET Journal
Application of ANFIS in Civil Engineering- A Critical ReviewApplication of ANFIS in Civil Engineering- A Critical Review
Application of ANFIS in Civil Engineering- A Critical Review
IRJET Journal6 views
A WORKSPACE SIMULATION FOR TAL TR-2 ARTICULATED ROBOT von IAEME Publication
A WORKSPACE SIMULATION FOR TAL TR-2 ARTICULATED ROBOT A WORKSPACE SIMULATION FOR TAL TR-2 ARTICULATED ROBOT
A WORKSPACE SIMULATION FOR TAL TR-2 ARTICULATED ROBOT
IAEME Publication176 views
IRJET- 3D Reconstruction of Surface Topography using Ultrasonic Transducer von IRJET Journal
IRJET- 3D Reconstruction of Surface Topography using Ultrasonic TransducerIRJET- 3D Reconstruction of Surface Topography using Ultrasonic Transducer
IRJET- 3D Reconstruction of Surface Topography using Ultrasonic Transducer
IRJET Journal12 views
Benchmarking of indoor localization and tracking systems (LTSs) von Kurata Takeshi
Benchmarking of indoor localization and tracking systems (LTSs)Benchmarking of indoor localization and tracking systems (LTSs)
Benchmarking of indoor localization and tracking systems (LTSs)
Kurata Takeshi308 views
292741121 gauge rr_for_an_optical_micrometer_industrial_type_machine von phgnome
292741121 gauge rr_for_an_optical_micrometer_industrial_type_machine292741121 gauge rr_for_an_optical_micrometer_industrial_type_machine
292741121 gauge rr_for_an_optical_micrometer_industrial_type_machine
phgnome96 views
TUW-ASE Summer 2015 - Quality of Result-aware data analytics von Hong-Linh Truong
TUW-ASE Summer 2015 - Quality of Result-aware data analyticsTUW-ASE Summer 2015 - Quality of Result-aware data analytics
TUW-ASE Summer 2015 - Quality of Result-aware data analytics
Hong-Linh Truong884 views
Supporting Change in Product Lines within the Context of Use Case-driven Deve... von Lionel Briand
Supporting Change in Product Lines within the Context of Use Case-driven Deve...Supporting Change in Product Lines within the Context of Use Case-driven Deve...
Supporting Change in Product Lines within the Context of Use Case-driven Deve...
Lionel Briand374 views
IRJET- Optimization of Machining Facility Layout by using Simulation: Cas... von IRJET Journal
IRJET-  	  Optimization of Machining Facility Layout by using Simulation: Cas...IRJET-  	  Optimization of Machining Facility Layout by using Simulation: Cas...
IRJET- Optimization of Machining Facility Layout by using Simulation: Cas...
IRJET Journal44 views
M.tech. mechanical engineering 2016 17 von Piyush Pant
M.tech. mechanical engineering 2016 17M.tech. mechanical engineering 2016 17
M.tech. mechanical engineering 2016 17
Piyush Pant312 views
A M ULTI -O BJECTIVE B ASED E VOLUTIONARY A LGORITHM AND S OCIAL N ETWOR... von IJCI JOURNAL
A M ULTI -O BJECTIVE  B ASED  E VOLUTIONARY  A LGORITHM AND  S OCIAL  N ETWOR...A M ULTI -O BJECTIVE  B ASED  E VOLUTIONARY  A LGORITHM AND  S OCIAL  N ETWOR...
A M ULTI -O BJECTIVE B ASED E VOLUTIONARY A LGORITHM AND S OCIAL N ETWOR...
IJCI JOURNAL186 views

Más de Kurata Takeshi

HARC: Human Augmentation Research Center von
HARC: Human Augmentation Research CenterHARC: Human Augmentation Research Center
HARC: Human Augmentation Research CenterKurata Takeshi
30 views4 Folien
SC 24でのメタバース関連標準化概要:ヘルスケア応用事例を交えて(ISO/IEC JTC 1/SC 24) von
SC 24でのメタバース関連標準化概要:ヘルスケア応用事例を交えて(ISO/IEC JTC 1/SC 24)SC 24でのメタバース関連標準化概要:ヘルスケア応用事例を交えて(ISO/IEC JTC 1/SC 24)
SC 24でのメタバース関連標準化概要:ヘルスケア応用事例を交えて(ISO/IEC JTC 1/SC 24)Kurata Takeshi
47 views32 Folien
Standards and projects of SC 24/WG 9 on Metaverse and Interverse von
Standards and projects of SC 24/WG 9 on Metaverse and InterverseStandards and projects of SC 24/WG 9 on Metaverse and Interverse
Standards and projects of SC 24/WG 9 on Metaverse and InterverseKurata Takeshi
79 views17 Folien
屋内測位技術の応用事例とPDRベンチマーク標準化委員会の活動概要 von
屋内測位技術の応用事例とPDRベンチマーク標準化委員会の活動概要屋内測位技術の応用事例とPDRベンチマーク標準化委員会の活動概要
屋内測位技術の応用事例とPDRベンチマーク標準化委員会の活動概要Kurata Takeshi
260 views53 Folien
作業エリア遷移モデル生成とそのクラスター分析に基づく製造ラインの作業分析 von
作業エリア遷移モデル生成とそのクラスター分析に基づく製造ラインの作業分析作業エリア遷移モデル生成とそのクラスター分析に基づく製造ラインの作業分析
作業エリア遷移モデル生成とそのクラスター分析に基づく製造ラインの作業分析Kurata Takeshi
102 views27 Folien
「遠隔リハビリのための多感覚XR-AI技術基盤構築と保健指導との互恵ケア連携」で目指すところ von
「遠隔リハビリのための多感覚XR-AI技術基盤構築と保健指導との互恵ケア連携」で目指すところ「遠隔リハビリのための多感覚XR-AI技術基盤構築と保健指導との互恵ケア連携」で目指すところ
「遠隔リハビリのための多感覚XR-AI技術基盤構築と保健指導との互恵ケア連携」で目指すところKurata Takeshi
206 views20 Folien

Más de Kurata Takeshi(20)

HARC: Human Augmentation Research Center von Kurata Takeshi
HARC: Human Augmentation Research CenterHARC: Human Augmentation Research Center
HARC: Human Augmentation Research Center
Kurata Takeshi30 views
SC 24でのメタバース関連標準化概要:ヘルスケア応用事例を交えて(ISO/IEC JTC 1/SC 24) von Kurata Takeshi
SC 24でのメタバース関連標準化概要:ヘルスケア応用事例を交えて(ISO/IEC JTC 1/SC 24)SC 24でのメタバース関連標準化概要:ヘルスケア応用事例を交えて(ISO/IEC JTC 1/SC 24)
SC 24でのメタバース関連標準化概要:ヘルスケア応用事例を交えて(ISO/IEC JTC 1/SC 24)
Kurata Takeshi47 views
Standards and projects of SC 24/WG 9 on Metaverse and Interverse von Kurata Takeshi
Standards and projects of SC 24/WG 9 on Metaverse and InterverseStandards and projects of SC 24/WG 9 on Metaverse and Interverse
Standards and projects of SC 24/WG 9 on Metaverse and Interverse
Kurata Takeshi79 views
屋内測位技術の応用事例とPDRベンチマーク標準化委員会の活動概要 von Kurata Takeshi
屋内測位技術の応用事例とPDRベンチマーク標準化委員会の活動概要屋内測位技術の応用事例とPDRベンチマーク標準化委員会の活動概要
屋内測位技術の応用事例とPDRベンチマーク標準化委員会の活動概要
Kurata Takeshi260 views
作業エリア遷移モデル生成とそのクラスター分析に基づく製造ラインの作業分析 von Kurata Takeshi
作業エリア遷移モデル生成とそのクラスター分析に基づく製造ラインの作業分析作業エリア遷移モデル生成とそのクラスター分析に基づく製造ラインの作業分析
作業エリア遷移モデル生成とそのクラスター分析に基づく製造ラインの作業分析
Kurata Takeshi102 views
「遠隔リハビリのための多感覚XR-AI技術基盤構築と保健指導との互恵ケア連携」で目指すところ von Kurata Takeshi
「遠隔リハビリのための多感覚XR-AI技術基盤構築と保健指導との互恵ケア連携」で目指すところ「遠隔リハビリのための多感覚XR-AI技術基盤構築と保健指導との互恵ケア連携」で目指すところ
「遠隔リハビリのための多感覚XR-AI技術基盤構築と保健指導との互恵ケア連携」で目指すところ
Kurata Takeshi206 views
遠隔リハビリ研究とオンライン学会運営から見たコミュニケーションDX von Kurata Takeshi
遠隔リハビリ研究とオンライン学会運営から見たコミュニケーションDX遠隔リハビリ研究とオンライン学会運営から見たコミュニケーションDX
遠隔リハビリ研究とオンライン学会運営から見たコミュニケーションDX
Kurata Takeshi37 views
国際標準化におけるAR/MR用語の使われ方 von Kurata Takeshi
国際標準化におけるAR/MR用語の使われ方国際標準化におけるAR/MR用語の使われ方
国際標準化におけるAR/MR用語の使われ方
Kurata Takeshi187 views
XRに基づく遠隔リハの研究・事業事例調査報告 von Kurata Takeshi
XRに基づく遠隔リハの研究・事業事例調査報告XRに基づく遠隔リハの研究・事業事例調査報告
XRに基づく遠隔リハの研究・事業事例調査報告
Kurata Takeshi514 views
「遠隔リハビリのための多感覚XR-AI技術基盤構築と保健指導との互恵ケア連携」の概要 von Kurata Takeshi
「遠隔リハビリのための多感覚XR-AI技術基盤構築と保健指導との互恵ケア連携」の概要「遠隔リハビリのための多感覚XR-AI技術基盤構築と保健指導との互恵ケア連携」の概要
「遠隔リハビリのための多感覚XR-AI技術基盤構築と保健指導との互恵ケア連携」の概要
Kurata Takeshi322 views
サービス学とか何か(応用サービス工学) von Kurata Takeshi
サービス学とか何か(応用サービス工学)サービス学とか何か(応用サービス工学)
サービス学とか何か(応用サービス工学)
Kurata Takeshi331 views
XR/xDRによる労働生産性の向上、QoW向上 von Kurata Takeshi
XR/xDRによる労働生産性の向上、QoW向上XR/xDRによる労働生産性の向上、QoW向上
XR/xDRによる労働生産性の向上、QoW向上
Kurata Takeshi1K views
地理空間インテリジェンス技術を用いた 製造ラインでの作業分析 von Kurata Takeshi
地理空間インテリジェンス技術を用いた 製造ラインでの作業分析地理空間インテリジェンス技術を用いた 製造ラインでの作業分析
地理空間インテリジェンス技術を用いた 製造ラインでの作業分析
Kurata Takeshi287 views
Geospatial Intelligence for Health and Productivity Management in Japanese Re... von Kurata Takeshi
Geospatial Intelligence for Health and Productivity Management in Japanese Re...Geospatial Intelligence for Health and Productivity Management in Japanese Re...
Geospatial Intelligence for Health and Productivity Management in Japanese Re...
Kurata Takeshi1.2K views
製造業・サービス業での人とシステムとの協調 von Kurata Takeshi
製造業・サービス業での人とシステムとの協調製造業・サービス業での人とシステムとの協調
製造業・サービス業での人とシステムとの協調
Kurata Takeshi1.2K views
地理空間インテリジェンス:屋内測位技術を用いた現場のラボ化に基づくサービス研究事例 von Kurata Takeshi
地理空間インテリジェンス:屋内測位技術を用いた現場のラボ化に基づくサービス研究事例地理空間インテリジェンス:屋内測位技術を用いた現場のラボ化に基づくサービス研究事例
地理空間インテリジェンス:屋内測位技術を用いた現場のラボ化に基づくサービス研究事例
Kurata Takeshi494 views
健康経営のための地理空間インテリジェンス(GSI)に関する一考察 von Kurata Takeshi
健康経営のための地理空間インテリジェンス(GSI)に関する一考察健康経営のための地理空間インテリジェンス(GSI)に関する一考察
健康経営のための地理空間インテリジェンス(GSI)に関する一考察
Kurata Takeshi595 views
Communication beyond spatiotemporal constraints von Kurata Takeshi
Communication beyond spatiotemporal constraintsCommunication beyond spatiotemporal constraints
Communication beyond spatiotemporal constraints
Kurata Takeshi584 views
応用サービス工学研究室紹介2020 von Kurata Takeshi
応用サービス工学研究室紹介2020応用サービス工学研究室紹介2020
応用サービス工学研究室紹介2020
Kurata Takeshi1.3K views
OTASCE Map: A Mobile Map Tool with Customizable Audio-Tactile Cues for the Vi... von Kurata Takeshi
OTASCE Map: A Mobile Map Tool with Customizable Audio-Tactile Cues for the Vi...OTASCE Map: A Mobile Map Tool with Customizable Audio-Tactile Cues for the Vi...
OTASCE Map: A Mobile Map Tool with Customizable Audio-Tactile Cues for the Vi...
Kurata Takeshi1.6K views

Último

02. COLEGIO - KIT SANITARIO .pdf von
02. COLEGIO - KIT SANITARIO .pdf02. COLEGIO - KIT SANITARIO .pdf
02. COLEGIO - KIT SANITARIO .pdfRAULALEJANDROMALDONA
5 views7 Folien
MODULE-1 CHAPTER 3- Operators - Object Oriented Programming with JAVA von
MODULE-1 CHAPTER 3- Operators - Object Oriented Programming with JAVAMODULE-1 CHAPTER 3- Operators - Object Oriented Programming with JAVA
MODULE-1 CHAPTER 3- Operators - Object Oriented Programming with JAVADemian Antony D'Mello
8 views14 Folien
Field Programmable Gate Arrays : Architecture von
Field Programmable Gate Arrays : ArchitectureField Programmable Gate Arrays : Architecture
Field Programmable Gate Arrays : ArchitectureUsha Mehta
33 views74 Folien
Solution Challenge Introduction.pptx von
Solution Challenge Introduction.pptxSolution Challenge Introduction.pptx
Solution Challenge Introduction.pptxGDSCCEC
13 views16 Folien
Trust Metric-Based Anomaly Detection via Deep Deterministic Policy Gradient R... von
Trust Metric-Based Anomaly Detection via Deep Deterministic Policy Gradient R...Trust Metric-Based Anomaly Detection via Deep Deterministic Policy Gradient R...
Trust Metric-Based Anomaly Detection via Deep Deterministic Policy Gradient R...IJCNCJournal
5 views25 Folien
Ansari: Practical experiences with an LLM-based Islamic Assistant von
Ansari: Practical experiences with an LLM-based Islamic AssistantAnsari: Practical experiences with an LLM-based Islamic Assistant
Ansari: Practical experiences with an LLM-based Islamic AssistantM Waleed Kadous
13 views29 Folien

Último(20)

MODULE-1 CHAPTER 3- Operators - Object Oriented Programming with JAVA von Demian Antony D'Mello
MODULE-1 CHAPTER 3- Operators - Object Oriented Programming with JAVAMODULE-1 CHAPTER 3- Operators - Object Oriented Programming with JAVA
MODULE-1 CHAPTER 3- Operators - Object Oriented Programming with JAVA
Field Programmable Gate Arrays : Architecture von Usha Mehta
Field Programmable Gate Arrays : ArchitectureField Programmable Gate Arrays : Architecture
Field Programmable Gate Arrays : Architecture
Usha Mehta33 views
Solution Challenge Introduction.pptx von GDSCCEC
Solution Challenge Introduction.pptxSolution Challenge Introduction.pptx
Solution Challenge Introduction.pptx
GDSCCEC13 views
Trust Metric-Based Anomaly Detection via Deep Deterministic Policy Gradient R... von IJCNCJournal
Trust Metric-Based Anomaly Detection via Deep Deterministic Policy Gradient R...Trust Metric-Based Anomaly Detection via Deep Deterministic Policy Gradient R...
Trust Metric-Based Anomaly Detection via Deep Deterministic Policy Gradient R...
IJCNCJournal5 views
Ansari: Practical experiences with an LLM-based Islamic Assistant von M Waleed Kadous
Ansari: Practical experiences with an LLM-based Islamic AssistantAnsari: Practical experiences with an LLM-based Islamic Assistant
Ansari: Practical experiences with an LLM-based Islamic Assistant
M Waleed Kadous13 views
Programmable Switches for Programmable Logic Devices von Usha Mehta
Programmable Switches for Programmable Logic DevicesProgrammable Switches for Programmable Logic Devices
Programmable Switches for Programmable Logic Devices
Usha Mehta37 views
AWS Certified Solutions Architect Associate Exam Guide_published .pdf von Kiran Kumar Malik
AWS Certified Solutions Architect Associate Exam Guide_published .pdfAWS Certified Solutions Architect Associate Exam Guide_published .pdf
AWS Certified Solutions Architect Associate Exam Guide_published .pdf
Basic Design Flow for Field Programmable Gate Arrays von Usha Mehta
Basic Design Flow for Field Programmable Gate ArraysBasic Design Flow for Field Programmable Gate Arrays
Basic Design Flow for Field Programmable Gate Arrays
Usha Mehta24 views
Unlocking Research Visibility.pdf von KhatirNaima
Unlocking Research Visibility.pdfUnlocking Research Visibility.pdf
Unlocking Research Visibility.pdf
KhatirNaima11 views
DevFest 2023 Daegu Speech_이재규, Implementing easy and simple chat with gol... von JQLEE6
DevFest 2023 Daegu Speech_이재규,  Implementing easy and simple chat with gol...DevFest 2023 Daegu Speech_이재규,  Implementing easy and simple chat with gol...
DevFest 2023 Daegu Speech_이재규, Implementing easy and simple chat with gol...
JQLEE616 views
Different type of computer networks .pptx von nazmul1514788
Different  type of computer networks .pptxDifferent  type of computer networks .pptx
Different type of computer networks .pptx
nazmul151478820 views

Work Pattern Analysis with and without Site-specific Information in a Manufacturing Line

  • 1. Work Pattern Analysis w/ & w/o Site-specific Information in a Manufacturing Line Takeshi Kurata1, Rei Watanabe2 Satoki Ogiso1, Ikue Mori1, Takahiro Miura1, Karimu Kato1 Yasunori Haga3, Shintaro Hatakeyama3, Atsushi Kimura3 and Katsuko Nakahira2 1 Human Augmentation Research Center, AIST, Japan 2 Faculty of Engineering, Nagaoka University of Technology, Japan 3 DENSO CORPORATION, Japan APMS 2023
  • 2. HPM and Geospatial Intelligence (GSI) 2 Kurata, T., Geospatial Intelligence for Health and Productivity Management in Japanese Restaurants and Other Industries, APMS, pp. 206–214 (2021) doi: 10.1007/978-3-030-85906-0_23
  • 3. Geospatial Intelligence (GSI) with 6M data 3 IE: Industrial Engineering, OR: Operations Research, IoT: Internet of Things, IoH: Internet of Humans, UI: User Interface XR: VR, AR, MR, etc. (AR: Augmented Reality, VR: Virtual Reality, MR: Mixed Reality) RM: Raw Material, WIP: Work In Progress, SFG: Semi-Finished Goods, FG: Finished Goods Conceptual diagram of GSI 6M information in Service and Manufacturing sites
  • 4. Study focus and research question 5 DENSO CORPORATION: Promoting activities aimed at improving both productivity and QoW, especially the working environment, such as the thermal environment. DENSO & AIST: Conducting multifaceted analysis of productivity and QoW using various acquired data while utilizing indoor GSI technology. Focus of this study (today’s presentation): Work analysis using flow line data for capturing a comprehensive view of the actual work behavior of each worker on the manufacturing line. Research question: Examine whether it is possible to start the analysis even before site-specific information is available. Room temperature heat map and flow line (exposed temperature and activity of each worker)
  • 5. Work pattern analysis w/ GSI: Start with or without site-specific information? 4
  • 6. Work site in this study 6 Manufacturing line in a DENSO factory • Manufacturing process of automotive work-in-progress • Main area for analysis: manufacturing line (1,400 m2) Measurement period: 5 days Number of workers: 10 • Day shift: 5 (Leader, Deputy leader, Receiving, Visual inspection, Internal inspection) • Night shift: 5 (Leader, Deputy leader, Receiving, Visual inspection, Assembly) Manufacturing line Break room Air shower Staying plots
  • 7. Manufacturing line Air shower Break room Indoor Positioning 7 Indoor positioning system in this study (PDR & BLE & Map) CE50: about 3m Indoor positioning tech map [16] S.Ogiso, et al., “Integration of BLE-based proximity detection with particle filter for day-long stability in workplaces,” IEEE/ION PLANS 2023 BLE beacons • 48 locations in the manufacturing line • 8 locations in the break room and others
  • 8. Overview of our work analysis method 8 Staying plots Extract “staying plots” from flow lines Generate the “work area transition model” by clustering of staying plots of all shifts & workers Obtain “work area transition instances” by registering features of each shift & worker in the work area transition model Find “work patterns” by clustering of work area transition instances Extract exceptions from work area transition instances by comparing to work patterns Work area transition model Clustering Work pattern A (Cluster A) Work pattern B (Cluster B) Work pattern C (Cluster C) Clustering Indoor positioning Work area transition instances Exception 1 (non-routine work 1) Exception 2 (non-routine work 2)
  • 9. How to generate the work area transition model 9 K-means++
  • 10. Work area transition model generated in this study 10 525 dimensions in total • 84-dimensional staying rate features • 441-dimensional moving rate features
  • 11. How to obtain a work area transition instance w/ the work area transition model 11 Registering staying rate features and moving rate features of each shift & worker in the work area transition model Staying rate features Moving rate features
  • 12. Extract “staying plots” from flow lines Generate the “work area transition model” by clustering of staying plots of all shifts & workers Obtain “work area transition instances” by registering features of each shift & worker in the work area transition model Find “work patterns” by clustering of work area transition instances Extract exceptions from work area transition instances by comparing to work patterns Find work patterns 12 Staying plots Exception 1 (non-routine work 1) Work area transition model Clustering Work pattern A (Cluster A) Work pattern B (Cluster B) Work pattern C (Cluster C) Clustering Indoor positioning Work area transition instances Exception 2 (non-routine work 2) K-means++
  • 13. Examples of work area transition instances and clustering results 13 Example of work area transition instances in the typical work patterns for each role Assumed deployment information for each role provided by the site manager after clustering Clustering results of work area transition instances and their relationship to roles • ID 1-8: Work pattern IDs (cluster IDs) • ID 0: Four work patterns with one instance • 46 work area transition instances into 12 clusters Note: Role and shift info are used only as ground truth (not for clustering). 1 2 3 5 7 8
  • 14. 14 How to extract exceptions (non-routine works) from work area transition instances RSS: Residual Sum of Squares
  • 15. Extracted exceptions of work area transition instances 15 Seven extracted exceptions of work area transition instances along with their exception indicators and the threshold. Exceptions A and B for visual inspection. 5
  • 16. Conclusions 16 12 work patterns and 7 non-routine works found w/o site-specific information containing each worker's role, shift and typical work areas of each worker More findings by interviewing on-site managers w/ the analysis results (Not discussed today...) Future works Verification of the applicability in other sites Pre-evaluation of site improvement/design ideas by simulation based on the work area transition model and work patterns Multifaceted analysis of productivity and QoW w/ physical-work data, operational data, environmental exposure data, vital sign data, and subjective data Work pattern analysis only w/ flow lines based on indoor GSI (Geospatial Intelligence)
  • 17. On-going other case: Express-way service area with a two- story building 17 26 clusters of staying plots (48 workers, 20 days) Work area transition model Examples of work area transition instances 1F 2F 1F 2F
  • 18. Evaluating the work environment and physical load of factory workers 18 [15] Nakae, S., el al., Geospatial intelligence system for evaluating the work environment and physical load of factory workers, 45th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), 5 pages, 2023 Multifaceted analysis of productivity and QoW w/ physical-work data, operational data, environmental exposure data, vital sign data, and subjective data
  • 19. Previous related work: Simulation with a work process model generated from flow lines and picking data in a warehouse 19 Single picking Zone picking Myokan, T., et al., Pre-evaluation of Kaizen plan considering efficiency and employee satisfaction by simulation using data assimilation -Toward constructing Kaizen support framework-, Proc. International Conference of Serviceology (ICServ2016), 7 pages (2016)
  • 20. Conclusions 20 12 work patterns and 7 non-routine works found w/o site-specific information containing each worker's role, shift and typical work areas of each worker More findings by interviewing on-site managers w/ the analysis results (Not discussed today...) Future works Verification of the applicability in other sites Pre-evaluation of site improvement/design ideas by simulation based on the work area transition model and work patterns Multifaceted analysis of productivity and QoW w/ physical-work data, operational data, environmental exposure data, vital sign data, and subjective data Work analysis only w/ flow lines based on indoor GSI (Geospatial Intelligence)