ANLY482 AY2017-18T2 Group19 Project Overview
Fujitsu-SMU Urban Computing and Engineering (UNiCEN) Corp. Lab is part of the Urban Computing & Engineering Centre of Excellence (CoE), a Public-Private Partnership between Agency for Science, Technology and Research (A*STAR), Fujitsu Limited (Fujitsu) and Singapore Management University (SMU) established on 15 Oct 2014. UNiCEN focuses mainly on two research areas, namely Dynamic Mobility Management (DMM) and Maritime & Port Optimisation (MPO). These two areas place emphasis on solving problems associated with crowds, congestion and queues, and resource management.
This project aims to provide a staff scheduling and route plan for nurses who are providing routine checkups and other services to the elderly in the comfort of their own homes. We aim to develop an algorithm, taking into consideration the appointment duration, traveling time to the patient’s location and other constraints to optimise the schedule for the nurses. This algorithm will be able to be used by all home health care foundations in the future to help them schedule their nurses efficiently. We will build a simple web application to allow the user to upload a csv file of the patients’ locations and prefered appointment time. The system will then run the algorithm which will give a detailed breakdown of the nurses’ schedules. The schedule will include the start and end time for a patient at a location.
Singapore is experiencing an aging population, experiencing an increase in median age from 40 to 40.5 years old and 12.4% to 13% for the population aged 65 and above in 2017. This results in an increase in difficulties that the elderly face, such as disabilities or being frail, in order for them to go to a hospital to have their checkup. This may cause discomfort and a less than satisfactory check-up throughout the whole process from going to and from the hospital and thus the need for home healthcare services are increasing. Health Minister Gan Kim Yong estimated an increase in 7600 more places in day, home and palliative care by 2020. The current situation may still be acceptable where nurses are given the cases and locations and they will plan it out themselves. However when the number of cases per nurse increases exponentially, the current system may not provide the most efficient services to the patients. By working on this project, we hope that we are able to maximise the level of satisfaction for the elderly during their check-ups while utilising the nurses efficiently by planning the shortest route to achieve the most number of services in a day.
1. Provide an algorithm for the scheduling of staffs that maximises the followings:
- Maximizing number of matches of prefered time slots from patients
- Maximizing number of revisiting nurses to patients instead of a new nurse each visit
- Maximizing utilisation of nurses in a day
2. Provide a UI for staffs to see their schedules
3. Provide dashboard to show the staff utilisation rate, plot out the route of the nurse for the day and inform where they should be at the current time
1. Data Preparation
- Data collection
- Data cleaning
- Exploration of data
2. Model Planning
- Problem Definition
- Generation of model objective and constraints
3. Model Building
- Development of algorithm
4. Model Evaluation
- Model validation
- Improvements in model score and per client’s recommendation
