Abstract—Our study is concerned with the development of an Emergency Medical Services (EMS) ambulance location and allocation model called the Time-based Ambulance Zoning Optimization Model (TAZ_OPT). This paper presents the framework of the study. The model is formulated using the goal programming (GP), where the goals are to determine the satellite locations of ambulances and the number of ambulances to be allocated at these locations. The model aims at maximizing the expected demand coverage based on probability of reaching the emergency location within targetted time, and minimizing the ambulance busyness likelihood value. Among the benefits of the model is the increased accessibility and availability of ambulances, thus, enhanced quality of the EMS ambulance services. Keywords—Optimization; Ambulance Location; Location facilities. I. INTRODUCTION MBULANCE is one of the EMS components and it is available 24 hours per day in most hospitals. The EMS ambulance services provide emergency care and transport patients to hospital immediately to reduce patients’ mortality, disability or suffering. The risk of death of patients requiring emergency treatment might increase due to long ambulance’s response time. Response time is defined by the time elapse between the minutes an operator finished receiving information from a caller to the time an ambulance arrives at the emergency site [1]. Since time is the main concern to those involved in the emergency, therefore EMS ambulance response time plays an important role to measure the quality of the EMS ambulances’ performance. Strategic ambulance location and allocation is a way to reduce the extensive response time. This paper presents the conceptual framework of our study, which concerns with the development of a mathematical model for ambulance location and allocation problem. This paper is organized as follows; I. Introduction, II. Literature review, III. Conceptual Research Framework, and IV. Summary. II. LITERATURE REVIEW Several studies regarding the ambulance location and allocation model can be found as early as in 1970s. One of the earliest models that can be found is the Location Set Covering Model (LSCM) introduced by Toregas et al. in 1971 [2]. LSCM determines the suitable ambulance locations to cover all demands within pre-specified time and distance while limiting the number of ambulances. Later in 1974, Church et al. [3] proposed the Maximum Coverage Location Problem (MCLP) to determine the EMS ambulances’ stations by maximizing demand covered with one condition that is limiting the number of ambulances allocated at these potential stations. Even though both models proposed by [2] and [3], respectively, are able to cover all demands, an important factor which is the ambulance busyness has been neglected. Here forth, various models were developed based on these two models, where some of them take into consideration the ambulance busyness as a parameter. Some extensions to the LSCM are the probabilistic LSCM (PLCSM) developed by ReVelle and Hogan [4], the Reliability model (Rel-P) by Ball and Lin [5] and the Ambulance Allocation Coverage Model (AACM) by Shiah and Chen [6]. The PLSCM [4] proposed the probability value of the server (ambulance) busyness constraint into the LSCM. This model was then extended to the Queuing PLSCM (QPLSCM) by Marianov and ReVelle [7]. Meanwhile, Rel-P [5] is capable of covering all demands using unlimited number of ambulances while considering the minimum cost to locate the ambulances at potential locations. On the other hand, the AACM [6] incorporates the capacity of ambulance service area into LSCM by considering the city’s physical road structure (single line road map) and population distribution (address point map). The Maximum Expected Covering Location Problem (MEXCLP) by Daskin [8], Maximum Availability Location Problem (MALP) by ReVelle and Hogan [9] and the GP model by Alsalloum and Rand [10] are some extension models for the MCLP. MEXCLP [8] considers the probability of server (ambulance) being busy when maximizing demand covered, where the probability was Framework of TAZ_OPT Model for Ambulance Location and Allocation Problem Adibah Shuib and Zati Aqmar Zaharudin A Adibah Shuib 1 , Zati Aqmar Zaharudin 2 are with the Faculty of Computer and Mathematical Sciences of the university Teknologi MARA, 40450 Shah Alam, Selangor, Malaysia (phone 1 : +603-55435307; fax 1 : +603-55435301; e- mail: adibah@tmsk.uitm.edu.my, zati_aqm@yahoo.com). World Academy of Science, Engineering and Technology 46 2010 677