Decision support for queue management systems - some reflections. Ana Respício University of Lisbon OR Center

Transcription

1 Decision support for queue management systems - some reflections Ana Respício University of Lisbon OR Center

2 Agenda Queue systems Motivation Problems/ sub-problems Envisaged system 2 IFIP TC8/WG8.3 Case Studies meeting LSE

3 Traditional queue NEXT! 3 IFIP TC8/WG8.3 Case Studies meeting LSE

4 Queue management system 1 A205 B090 C E-caller Ticket Assistant A205 1 B090 4 C020 5 A235 Service A ticket queue ticket dispenser 4 IFIP TC8/WG8.3 Case Studies meeting LSE

5 Queue management system A multiplicity of queues are created. Services are differentiated by letters allowing people to approach the right assistant. Modern systems for very crowded organizations / services already send text to phone giving customers information about the queue state, or Web publish it. 5 IFIP TC8/WG8.3 Case Studies meeting LSE

6 Image and efficiency Modernity and efficiency. Customer satisfaction customers can use waiting time doing other activities rather than just waiting. Sense of better quality of service. 6 IFIP TC8/WG8.3 Case Studies meeting LSE

7 Events A205 calling End/ type of service C020 1 Database 5 arrival A235 Service A 7 IFIP TC8/WG8.3 Case Studies meeting LSE

8 Data collected For each customer: Arrival time, Starting service time, Ending service time, Type of service/sub-services, Assistant serving. 8 IFIP TC8/WG8.3 Case Studies meeting LSE

9 Statistics Number of customers. Number of customers for each service. Number of customers giving up before service. Waiting time, service time. Sales (services and sub-services). Serving times for assistant/service. 9 IFIP TC8/WG8.3 Case Studies meeting LSE

10 Performance management Evaluating publicity campaigns. Quality of service evaluation. Evaluating/comparing stores performance. Human resources efficacy evaluation. 10 IFIP TC8/WG8.3 Case Studies meeting LSE

11 Decision support 1. Data mining Analysing correlations. Extracting tendencies. 2. Forecasting Making aggregate predictions for service requests over medium term planning period, e.g, a month. Making predictions for service requests over a short planning period, e.g., a week. 11 IFIP TC8/WG8.3 Case Studies meeting LSE

12 Decision support 3. Planning assistance counters. 4. Planning human-resources. 5. Real time decision support Monitoring and providing alerts and suggestions. 12 IFIP TC8/WG8.3 Case Studies meeting LSE

13 Planning assistance counters/ human-resources Scheduling workers Defining how many assistants should be assigned to each service for each daily working period (shifts) morning, evening, night during a given period of time, e.g., a week, ensuring a good quality of service. 13 IFIP TC8/WG8.3 Case Studies meeting LSE

14 Planning human-resources Rostering Defining which assistants should be assigned to each service at each shift, during a given period, e.g., a week, complying with legal and institutional rules, namely Labour Law, labour agreements and the company s regulations. providing a equitable distribution of work, at lowest cost. 14 IFIP TC8/WG8.3 Case Studies meeting LSE

15 Planning human-resources Re-scheduling/re-rostering Re-assigning assistants to services due to absences. Re-assigning assistants to services due to unexpected increasement or decreasement of arrivals. 15 IFIP TC8/WG8.3 Case Studies meeting LSE

16 Analysis/Planning/Scheduling Queue System Database Data Analysis Forecasting Decision Analysis Simulation Planning/ Scheduling 16 IFIP TC8/WG8.3 Case Studies meeting LSE

17 Real time decision support Defining metrics (KPI) to be monitorized and used to generate alerts Maximum/average people in queue Maximum/average waiting time for each service/assistant Maximum/average people giving up 17 IFIP TC8/WG8.3 Case Studies meeting LSE

18 Real time decision support Queue System Database Monitoring System Decision Analysis Simulation Heuristic Search 18 IFIP TC8/WG8.3 Case Studies meeting LSE

19 Discussion... Thank you 19 IFIP TC8/WG8.3 Case Studies meeting LSE