| Author: | Wong, Chi Wai |
| Title: | Design and implementation of a new computer vision system for enhancing the management efficiency of inventory control |
| Advisors: | Man, H. C. (ISE) Lam, Kenneth (EIE) |
| Degree: | Eng.D. |
| Year: | 2020 |
| Subject: | Inventory control Business logistics Warehouses -- Management Hong Kong Polytechnic University -- Dissertations |
| Department: | Department of Industrial and Systems Engineering |
| Pages: | 215 pages : color illustrations |
| Language: | English |
| Abstract: | Owing to the change of customer behavior, the total number of orders is increasing while the order quantity per order is decreasing. Most of the time, the minimum order quantity cannot be reached. In order to maximize the production effectiveness without disappointing the customer, the production quantity will be decided to be produced based on minimum order quantity which results in excessive units more than required units of the actual order. The excessive quantity will then be kept in warehouse as stock. As the accumulating stock reaches a high level, a proper warehouse management system will have to be built. Some companies will assign a third-party logistics company as they are the specialist in this area and incur less investment and human resources. Unfortunately, our customers do not allow us to subscribe the services from third-party logistics as they require a full control of supply chain. In other words, the finished goods can only be stored in our warehouse. The warehouse team needs to tackle the inventory management system on its own by setting up a comprehensive inventory control system. The system needs to provide (1) storage management, (2) inventory allocation, (3) inventory tracking and (4) inventory control. The inventory tracking is the most difficult task to be achieved in a giant warehouse. However, it is essential since it can provide useful information for inventory allocation and control. The most commonly used inventory tracking technology is using One-dimensional and Two-dimensional Barcode technology as it is the most cost-effective approach, but it is quite passive. The users need to bring the mobile barcode reader to scan the barcode attached onto the finished goods and the data will be sent to server for further processing. Although the barcode can be captured by the camera, the resolution of the image will affect the accuracy of decoding. In addition, most of the barcode application or software focus on one-to-one barcode decoding. The image contains several barcodes cannot be decoded without software or system customization. Another technology, RFID, will be worked in a more proactive approach which demonstrates many successful cases in supply chain. By coupling, the data from the tag can be sent to the reader in a wire speed. Moreover, the reader can simultaneously read hundreds of tags. However, the cost for inventory tracking is relatively expensive. Normally, RFID inventory tracking system will only keep track of the deluxe goods and expensive assets. In our situation, the monetary value of the finished goods on a pallet are around several hundred dollars. The return of investment is very low if applying RFID as the cost of active tag is around thousand dollars. Therefore, we shall look for a cost-effective approach. To complete inventory tracking, unique identification is most critical. After extensive research and experimental work, a SMART tag system has been designed and built. The SMART tag system combines the advantages of one-dimensional and two-dimensional barcode. With SMART tag, a new computer vision system is designed for warehouse management. The architecture of the computer vision system is extensible and scalable. It is designed to handle real-time and high-volume processing. The computer vision system not only provides the basic features of inventory control system, but also the advanced features, such as visual management and error detections, to facilitate warehouse users for daily operation and planning. During the pilot run, the computer vision system demonstrated how the system could solve the existing problems. The company decided to implement this innovative system as it could solve a major and critical problem of the company. The new system has the following innovative areas that win over traditional tracking systems: 1.) The design of Smart Tag facilitates to develop an algorithm to detect whether it exists in the image even if it cannot be decoded owing to distortion. This is different from the traditional system such as RFID. The traditional system is unable to alert users if it cannot read the tag. It is incapable of detecting the existence of the tag. 2.) Inspired by the architecture of Microservices and principle of separation of concerns, the architecture of the newly designed computer vision system can achieve the real-time requirements. In addition, it can seamlessly integrate with existing systems such as ERP. And, it is ready to provide services through internet in a secure manner. 3.) A newly designed protocol, named Fast Transfer, to transfer file fast and secure than the traditional File Transfer Protocol (FTP), can transfer the image files to the system for users to conduct visual management. Users do not need to go to warehouse to carry out visual management. This saves a lot of time. 4.) The cost of the newly designed system is much cheaper than the traditional system such as RFID to achieve inventory tracking. 5.) The architecture of the designed system is scalable and extensible. 6.) Although LED identification tag cannot be applied in warehouse efficiently, it is very suitable in a closed environment such as container. The invention of LED tag can be applied in other areas. |
| Rights: | All rights reserved |
| Access: | restricted access |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 6538.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 8.47 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/12099