This project started as SAP SMB Innovation Labs – IE Select APJ 2019 project. It started as simple face recognition and the project then evolved into something bigger. We integrate face recognition technology with our Trolley Point of Sales (POS) and our Customer Royalty Program. Finally, we decided to submit this solution for SME SEEDx Development Challenge 2020.
Read more about the challenge here.
BeOne Trolley Demo Video:
https://youtu.be/N3ZWNyQZx8o
Solution Name:
BeOne Trolley
Solution Description:
Ensuring, improving, and maintaining customer’s satisfaction, engagement, and retention are some of common problems found in retail industries. Often brands are dying because they do not know their customers well or they are only selling their products without good marketing strategy or doing it incorrectly.
Realizing this, we incorporated face recognition technology with our Trolley Point of Sales (POS). The face recognition system will detect existing customers, gold members or VIP customers, and even new customers. Cashier, front desk, or sales person can engage them personally, resulting in better customer experience. Furthermore, the application can give insights of customer’s transaction history, estimated age, gender, and expression information to be analysed, resulting in better customer’s satisfaction management. By knowing the customers well, utilizing personalized promotion strategy, sales division can maintain a better customer’s retention.
Solution Use Case:
The application is targeted at retail or food and beverage (FnB) stores. The face recognition module is a web-based application. Using camera (or simply webcam) attached to POS machine and network connection, the application is ready to go. Basically, the application will do the followings:
Persona Identified:
Pain Points:
Solution Details:
BeOne Trolley is consisted of some core modular systems:
Solution Technology:
UX Technology Used: HTML5, CSS3
Platform Technology Used: Beone Cloud Server
Latest Technology Used:
Framework Used:
Database:
Algorithm:
Industry Focus: Retail Industry and Food and Beverages (FnB)
Marketing Strategy:
Road Map:
Future development will include:
Contact Information:
Partner: PT. Beone Optima Solusi
Country: Indonesia
Team Name: BeOne Intelligent Enterprise Team
Team Members: Ronnie Eko Prasetyo, Oki Wijaya, Wirianta, Yoseph Sunarli, Reza Stefano, Brian Billardo
The Challenge is a big chance for us, first participating partner from Indonesia, to begin researching, investing, and exploring the fields of Intelligent Enterprise and Industry 4.0. We are really excited to keep submitting new solutions as we believe a lot of wild ideas about future technology and innovation can be delivered into real things starting right now.
For more information about the development challenge, you may refer to the SME SEEDx Development Challenge 2020.
Read more about the challenge here.
BeOne Trolley Demo Video:
https://youtu.be/N3ZWNyQZx8o
Submission Details
Solution Name:
BeOne Trolley
Solution Description:
Ensuring, improving, and maintaining customer’s satisfaction, engagement, and retention are some of common problems found in retail industries. Often brands are dying because they do not know their customers well or they are only selling their products without good marketing strategy or doing it incorrectly.
Realizing this, we incorporated face recognition technology with our Trolley Point of Sales (POS). The face recognition system will detect existing customers, gold members or VIP customers, and even new customers. Cashier, front desk, or sales person can engage them personally, resulting in better customer experience. Furthermore, the application can give insights of customer’s transaction history, estimated age, gender, and expression information to be analysed, resulting in better customer’s satisfaction management. By knowing the customers well, utilizing personalized promotion strategy, sales division can maintain a better customer’s retention.
Solution Use Case:
The application is targeted at retail or food and beverage (FnB) stores. The face recognition module is a web-based application. Using camera (or simply webcam) attached to POS machine and network connection, the application is ready to go. Basically, the application will do the followings:
- Capture real time customer’s image from webcam video feed.
- Recognize customer and display recognized name, estimated age, gender, and expression.
- Retrieve customer’s transaction history.
- Normal transactions using Trolley POS.
- Provide dashboard of all detected customers data.
- Provide a follow up module for any negative sentiments (expressions) which will create Service Call document on SAP Business One.
Persona Identified:
Pain Points:
Solution Details:
BeOne Trolley is consisted of some core modular systems:
- Face Recognition from webcam or camera live feed. It is a web application and can run on any browser as long as the Javascript is enabled. The web application will access the page hosted on our cloud server. It can fetch name, age, gender, and expression information using deep learning AI. Unrecognised name will be treated as Unknown and can be registered as new customer to the database via the Trolley POS.
- Trolley Point of Sales (POS) is our POS application running on Android platform. The POS can also run on POS Machine (for example, Sunmi).
- Dedicated Detection Scheduled Service. The service runs in the background on our cloud server to store every detected face every set time. The face recognition will communicate with this service periodically using web socket. This way, the face recognition system can run uninterrupted and computer resources can be fully allocated.
- Dashboard for analytical process. The dashboard will show data visually using tables and charts. Administrator can follow up any negative data (sad, angry expression) to create Service Call on SAP Business One from the dashboard via Service Call.
Solution Technology:
UX Technology Used: HTML5, CSS3
Platform Technology Used: Beone Cloud Server
Latest Technology Used:
- Face Recognition
- Machine Learning and Deep Learning
Framework Used:
- Node JS
- Face API JS
Database:
- MySQL
- SAP HANA
Algorithm:
- Single Shot Multibox Detector based on Mobile Net (SSD Mobilenet)
- Tiny Face Detector (for tinier version of SSD Mobilenet)
Go-to-Market Strategy
Industry Focus: Retail Industry and Food and Beverages (FnB)
Marketing Strategy:
- Social Media Marketing
- Monthly Workshop
- Beone Channel Partners
Road Map:
Future development will include:
- Gift Voucher with personal budgeting.
- Automatically sends personalised marketing campaign and promotion, customer feedback form (for any negative expression detected) through messaging service (i.e. WhatsApp, Line, Facebook Messenger, SMS, etc)
Contact Information:
Partner: PT. Beone Optima Solusi
Country: Indonesia
Team Name: BeOne Intelligent Enterprise Team
Team Members: Ronnie Eko Prasetyo, Oki Wijaya, Wirianta, Yoseph Sunarli, Reza Stefano, Brian Billardo
The Challenge is a big chance for us, first participating partner from Indonesia, to begin researching, investing, and exploring the fields of Intelligent Enterprise and Industry 4.0. We are really excited to keep submitting new solutions as we believe a lot of wild ideas about future technology and innovation can be delivered into real things starting right now.
For more information about the development challenge, you may refer to the SME SEEDx Development Challenge 2020.
- SAP Managed Tags:
