Improving water meter reading operations using optimization techniques for Syarikat Air Johor
Water meter distribution network is a challenge for decision makers to choose the best route during water bill distribution to ensure all places are visited without skipping any consumer’s house. In particular, rural areas are more challenging as compared to urban areas because there are a lot of ba...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/id/eprint/81487/1/NurhasyimahMohamadAliMFKM2015.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Water meter distribution network is a challenge for decision makers to choose the best route during water bill distribution to ensure all places are visited without skipping any consumer’s house. In particular, rural areas are more challenging as compared to urban areas because there are a lot of barriers or challenges faced by meter reader during water bill reading process. Based on observation and interview session there are four factors causing the problem during meter reading process which are geographic factor, human factor, machine factor and external factor. Furthermore, meter reading department needs to complete their task within a specific time frame which is in between 29 to 30 days for all consumers (58587 registered accounts). If billing days are less than 29 days the effect is profit to Syarikat Air Johor (SAJ), if more than 30 days SAJ will lose the profit. Win-win situation for SAJ and customers will happen if the billing days are exactly 30 days. Currently SAJ Kota Tinggi is unable to comply with this requirement. Overdue cycle is a financial lost to the company and early reading is a financial lost to the customers. The purpose of this study is to propose an optimum assignment and route for meter readers. Technique based on travelling salesman problem and assignment problem were used in this study. Evaluations were made using travelling salesman problem method and minimum spanning tree as a comparison for total time, sequence path and estimation time analysis. The findings suggest that the travelling salesman problem provided a minimum total time compared with minimum spanning tree and the current practice for both rural and urban areas covered in this study. In terms of estimated reading time, the result suggests 99% probability that meter reading task can be completed in 205 minutes for 175 rural areas consumers and 126 minutes for 167 urban areas consumers. |
|---|