Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1598-7248 (Print)
ISSN : 2234-6473 (Online)
Industrial Engineering & Management Systems Vol.17 No.1 pp.43-61

A Process Reference Model for Hospital Supply Chain of Pharmaceutical Products

Wirachchaya Chanpuypetch*, Duangpun Kritchanchai
Department of Industrial Engineering, Faculty of Engineering, Mahidol University, Nakhonpathom, Thailand
Corresponding Author,
March 8, 2017 May 8, 2017 June 12, 2017


An implementation of business process reengineering (BPR) is more complex in healthcare than other industries. Interdepartmental collaboration and acceptance of change are essential requirements. To achieve success, a prerequisite tool should be provided in the form of a process reference model (PRM). The structure of a competent PRM that includes necessary components is firstly proposed by consideration of the crucial requirements for BPR in healthcare systems. According to the proposed PRM structure, a Hospital supply chain PROcess reference Model of pharmaceutical products (H-PROM) has been constructed through an empirically based method. The heart of H-PROM is presented in this paper. Initially, a process reference framework is framed as a skeleton of the model. Then the core hospital tasks are defined and described in detail. Furthermore, these reference tasks are also represented in the form of a standard business process diagram using business process modelling notation (BPMN). These contributions can benefit a process engineer or an organisation from the developed PRM, in order to quickly understand the hospital supply chain system and accept change. Besides, this approach can be a way to propose the key ingredient for facilitating BPR in other healthcare processes.



    Healthcare is an important service industry in the economic industrial sector in Thailand. It is growing quickly and the country has become a medical hub for Asia. However, the Thai healthcare industry is still suffering from inefficient processes, and inconsistent and inaccurate information (Kritchanchai and Krichanchai, 2010). Based on this point, the authors conducted a preliminary empirical investigation to understand in-depth the current situation (AS-IS) in healthcare supply chain through hospital case studies in Thailand. Pharmaceutical product management in hospitals has formed a focus, as it is a critical part of the healthcare system and its stocks need to be guaranteed. As a result of the preliminary analysis, the researchers found that the management of pharmaceutical products within hospitals involves a variety of business processes and information systems. Multiple computerised information systems that relate to pharmaceutical product management have been implemented in hospitals. However, a few hospitals manage their inventory on manually based information systems, such as stock cards. Besides, Thai hospitals allow each department (procurement, hospital warehouse, hospital pharmacy, and wards) to operate their activities independently. Hence, a lack of both continuity and information sharing occurs, often known as a fragmented system, in the hospital supply chain of pharmaceutical products. Due to this situation, a strong bias is often generated and brings about inaccurate forecasts, high inventory levels, and high costs of inventory management. Moreover, if physicians make a prescription by hand, inventory and drug information cannot be shared with them. Drugs that are not stored within the hospital may be prescribed. Medication errors and process delays are unavoidable. As mentioned, based on empirical evidence, various problems can be found with respect to fragmented process structures.

    According to this AS-IS situation, business process reengineering (BPR) has been proposed as a solution for eliminating a functional approach for providing connected processes (Hammer and Champy, 1993). It has been considered as an important way to reshape business organisations for achieving breakthrough improvements in performance (Wu, 2002). Redundant processes and delays can be removed via BPR approaches (Hammer and Champy, 1993; Patwardhan and Patwardhan, 2008). Thus, BPR may be a key tool for improving hospital supply chains in Thailand.

    Thailand. To carry out a BPR project, process engineers must understand the existing processes before concentrating on “what should be” (Hammer and Champy, 1993; Wu, 2002). Knowledge of the current behaviour of systems, business processes and players is an essential prerequisite for the implementation of any BPR project (Bertolini et al., 2011). Nevertheless, as mentioned previously, a variety of business processes have been found in Thai hospital supply chains of pharmaceutical products. Besides, many researchers note that healthcare is significantly more complex than other industries (El Farouk Imane et al., 2011; Rebuge and Ferreira, 2012). A high degree of collaboration among individuals and functions is often required in healthcare organisations (McConnell, 2010). Also, it is difficult to perform business process analysis in this context. A process engineer must take a long time to explicitly represent and thoroughly understand the current processes in a hospital system. It is inefficient if the endto- end business process must be renewed every time (La Rosa et al., 2008).

    Regarding this issue, a standard view of business processes should be offered. Nevertheless, this perspective is still not presented in the hospital supply chain context at this time (Kriegel et al., 2013; Nachtmann and Pohl, 2009). La Rosa et al. (2008) and Hertz et al. (2011) note that a lack of standardised form of process model or process orientation for supporting a BPR project can be proposed in the form of a process reference model (PRM). Hereby, this paper provides the PRM as an input ingredient for facilitating BPR in a hospital supply chain.

    This paper is organised as follows. Background and related works are provided in section 2. A construction method and the related foundations for reference modelling are then described in section 3. Section 4 introduces some components of the developed PRM including the process reference framework and a formalisation of modelling language. The main outcomes are discussed and concluded in section 5. Finally, further developments are discussed.


    2.1 Process Reference Model

    Recently, PRMs have gained increasing attention in the field of business process management. It is a specific kind of reference model that focuses on the behavioural aspects of an organisation through the analysis of its business processes (Blecken, 2010). The basis of organisations or companies for designing their processes in giving guidance on model content is provided as a documentation template of process knowledge (Scheer and Nüttgens, 2000). This approach can assist in business process management more efficiently (Bask et al., 2011).

    Continually, PRMs have been specifically proposed in various business settings such as event management companies (Thomas et al., 2008), the fruit industry (Verdouw et al., 2010), humanitarian organisations (Blecken, 2010), universities (Svensson and Hvolby, 2012), etc. These models can bring about various benefits such as being a standard tool for communication and collaboration to improve business systems (Carpinetti et al., 2003; Blecken, 2010; La Rosa et al., 2008), supporting of information system design and implementation (Thomas et al., 2008; Verdouw et al., 2010; Svensson and Hvolby, 2012), and being a tool for achieving change and handling cross-organisation processes (Molina et al., 2005; Hertz et al., 2011; Rabe et al., 2006; Svensson and Hvolby, 2012).

    Currently, these aspects are still not presented in the hospital supply chain (Kriegel et al., 2013; Nachtmann and Pohl, 2009). Therefore, the suitable components of PRM in this setting should be proposed as an input ingredient for facilitating BPR.

    2.2 BPR in Healthcare Environments

    In terms of healthcare, this system has the same characteristics as industrial environments, although with much more complexity (El Farouk Imane et al., 2011; McConnell, 2010). Accordingly, BPR projects have often been unsuccessful in their implementation (Patwardhan and Patwardhan, 2008). To achieve success, relevant literature was reviewed in order to understand the main issues and requirements that must be dealt with if BPR is to be successfully implemented in healthcare settings.

    As previous publications suggest, there are four main points that can be recognised in terms of hospital supply chains. First, healthcare professionals are the key players in the hospital supply chain (Khodambashi, 2013). Thus, an involvement of these staff is critical (Rebuge and Ferreira, 2012; Khodambashi, 2013). Second, the ultimate goal of healthcare systems is safety. This issue is more important than organisation needs (Patwardhan and Patwardhan, 2008). Accordingly, everyone in the organisation has to focus on a patient as the primary motivator (Francis and Alley, 1996). Third, the characteristic of healthcare systems is an interdepartmental organisation (McConnell, 2010). Therefore, a high degree of collaboration among individuals and functions with holistic approaches should be provided. Finally, resistance to change often exists when healthcare systems are redesigned (Patwardhan and Patwardhan, 2008; Khodambashi, 2013). The information necessary to manage the change process is often extremely lacking in healthcare organisations (Golden, 2006). This change requires supporting tools to ensure participation and acceptance of changes (Al-Abri, 2007; McConnell, 2010). A continuous change in complex healthcare environments is a challenge (MacPhee, 2007). Consequently, these requirements should be taken into consideration in the development of PRM in this context.

    2.3 PRM for Healthcare BPR

    As the requirements for BPR in healthcare described in the previous section, a holistic approach is essential for facilitating interdepartmental communication and collaboration. In addition, supporting tools should also be offered for improving participation and acceptance of change. Thus, the necessary elements which meet these requirements are then determined, by considering related literature detailing successful installation of PRMs with similar purposes.

    Even though many different purposes of the developed PRMs can be served, the three basic components are commonly found in any specific setting. They consist of a process reference framework, a document of the detailed model, and a PRM application. However, each component can establish various contributions. Initially, a process reference framework is constructed as a navigation frame, in order to develop the PRM for any purpose. Then, the presented models include the definitions and general descriptions as a document of the detailed model. A document can also present overall workflows by using a modelling language technique such as BPMN, EPC, or IDEF0. Additionally, some researchers encompass a description of design patterns (TO-BE) and a set of performance indicators into the model for being a supportable instrument for achieving change and handling cross-organisational processes (Rabe et al., 2006; Molina et al., 2005; Hertz et al., 2011; Svensson and Hvolby, 2012). For the application of PRM, most studies outline a basis structure for business stakeholders’ understanding to facilitate communication and collaboration by using a modelling language. Various contributions have been provided to support the analysis of processes and achieve change management, such as a configurable model (La Rosa et al., 2008; Verdouw et al., 2010; Molina et al., 2005; Svensson and Hvolby, 2012), exemplifying or demonstration of the provided PRM through a case study (Blecken, 2010; Reichert et al., 2013), or with a template referring to simulation modelling (Rabe et al., 2006; Molina et al., 2005). Furthermore, the procedure for model implementation can be offered for describing an application of the developed PRM (Carpinetti et al., 2003).

    Accordingly, the core components of PRM can be depicted in the following:

    • First, a process reference framework is initially constructed in order to frame the overall view of healthcare supply chain operation processes in the high-level of standard structure.

    • Second, based on the created framework, a document of the detailed model contains the general descriptions, the related business process diagram using modelling language techniques, a description of design patterns (TO-BE), and a set of performance indicators. Moreover, other details acquired from a real-life context can also be included.

    • Third, an application of PRM is created as a configurable model. It can serve as a blueprint for business analysts, domain experts, or organizations in order to derive the model for the specific organisation or the project. Also, this provided configurable model helps to maintain the complex set of healthcare business processes. Besides, templates referring to simulation modelling should be arranged in order to present a way for business process re-design and lead to acceptance of changes (Kerley et al., 2011). Also, the implementation steps of the developed PRM are described in order to guide users for applying the model.

    Concisely, the components of PRM for healthcare BPR should be composed of these aspects in order to achieve success. They are illustrated in Figure 1.


    In developing a new PRM, an empirically grounded PRM construction is recommended as the most suitable method, although a reference model can be developed theoretically or utilize document process know-how (Scheer and Nüttgens, 2000). Empirical evidence should be linked as the element of a reference model (O’Leary and Richardson, 2012). It is acquired as an important source of business process knowledge (Blecken, 2010; Galster and Avgeriou, 2011). A solid foundation for the PRM can be ensured through this approach. Nevertheless, there is no standard detailed description for construction steps. They can be conducted based on the problem that the PRM is attempting to solve (O’Leary and Richardson, 2012). Hence, the empirically based construction method for developing a Hospital supply chain PROcess reference Model (H-PROM) is proposed in Figure 2.

    As Figure 2 illustrates, a PRM can be developed via four stages. In this paper, they will be elaborated, focussing on the hospital supply chain context of pharmaceutical products. This hospital supply is crucial as their stock needs to be guaranteed (Schwarting et al., 2011). Besides, drug expenditure is an important factor in the profit and loss accounts of healthcare systems (Iannone et al., 2013). The four stages of an empirically based PRM construction are the following:

    Stage 1: Evaluation of reference modelling foundations

    Typically, good PRMs are not developed from scratch but based on generic principles and models with related theory regarding foundations for description, ordering and comparing to acquire knowledge of the context (Thomas et al., 2008). Therefore, the relevant foundations for reference modelling is firstly evaluated. The appropriate foundations that were selected based on the purposes and scope of the PRM construction are described in the following:

    Frameworks and reference models for supply chain management

    In this work, the Global Supply Chain Forum (GSCF) framework is suitable for framing an effectively navigable directory at the initial stage. It is one of the most widely known reference frameworks for SCM, developed by Cooper et al. (1997). This framework emphasises key characteristics of SCM, that is coordination of operation tasks and crucial SCM business processes within and between organisations in the supply chain (Stavrulaki and Davis, 2010). The eight key SCM business processes that all organisations in each supply chain should consider include customer relationship management, customer service management, demand management, order fulfilment, manufacturing flow management, supplier relationship management, product development and commercialisation, and return management. The framework breaks these crucial processes that cut across organisation and functional silos within each organisation from end user through to original suppliers (Stavrulaki and Davis, 2010). The entire SCM business process contains its own set of operational tasks that resides inside a functional silo (Croxton et al., 2001). A set of activities is performed under each task (Lambert et al., 2005). Accordingly, it can facilitate a cross-functional or cross-enterprise view of the organisation and its supply chain partners (Lambert et al., 2005; Ponis et al., 2013). A collaboration among individuals and functions can be displayed through the framework (Lambert et al., 2005) that illustrates a need for hospital system improvement (McConnell, 2010). Moreover, the format of GSCF is easy to understand. The standardisation for facilitating interdepartmental communication and collaboration can be achieved based on GSCF.

    The additional SCOR model is a reference model that was developed by the Supply-Chain Council (SCC) to provide a standard method for measuring supply chain performance and to create a common set of metrics that could be used for benchmarking purposes. This reference model is highly process-focused but the model cannot describe every business process or activity, including integrated cross-functional areas such as sales and marketing (demand generation) (Lambert et al., 2005; Ponis et al., 2013). Moreover, it is complex model and therefore requires a significant amount of training to understand and build the custom model. However, the SCOR model is a reference model that links process descriptions and definitions with metrics, best practice and technology (Ponis et al., 2013). Accordingly, some standardised elements of SCOR that may relate to hospital processes can be used to allow for a documentation of the detailed model.s

    Modelling language technique

    Business process modelling is the activity of representing processes of an enterprise (Chinosi and Trombetta, 2012). A good picture flow of activities that make up the key business processes in an organisation needs to be examined as an ingredient in successful redesign projects (Kock et al., 2009). Consequently, a modelling language technique, BPMN, is considered for representing hospital processes. BPMN is widely used as the inter-sectoral standard for process description, analysis and simulation (Correia and Abreu, 2012; Blecken, 2010). This notation is not only to provide easily understandable for technical developers, but also for business analysts and other nontechnical stakeholders (Khabbazi et al., 2013). Additionally, it is a convenient description technique for documenting and BPR (Recker et al., 2006). The key players for BPR in healthcare supply chains can perceive all their related tasks through a collaboration view that is created by using BPMN diagrams.

    Stage 2: Empirically based construction of PRM

    According to an empirically based construction method, a PRM will be developed in real-world situations by gathering empirical data (Scheer and Nüttgens, 2000). The data can be collected based on the SCM framework selected in the first stage by interview (Verdouw et al., 2010; Blecken, 2010). Other data that relate to the aim of model development are also gathered such as the details of activities, and AS-IS problems.

    Accordingly, empirical data related to pharmaceutical product management in a hospital are required in order to develop H-PROM. The four cases of a large hospital providing tertiary care and a teaching hospital where convenience to access in Thailand were selected for this research. These cases cover the phenomenon of the context of interest. The case study data were collected in semi-structured interviews using open-ended questions that were based on the GSCF framework. The interview pursued several objectives, including identifying the hospital functional silos; recognising the core tasks in each function area and the key SCM business processes of a hospital, perceiving phenomena or details of hospital tasks, and identifying problems of hospital supply chains in terms of pharmaceutical products. Those managers at senior level who could provide information on relevant business processes were interviewed. They included the head of a hospital pharmacy department, a pharmacy purchasing manager, a pharmacy warehouse manager, and a pharmacy store manager. All interviewees were asked the same questions, and interview length ranged from 1 to 3 hours. Simultaneously, direct observation was also performed.

    Consequently, the process reference framework can be constructed as a navigation frame. The definitions and descriptions of each task in the developed framework are then annotated based on the empirical evidence, generic principles, and the related reference framework and reference model for SCM.

    Besides, to build a robust model, qualitative analysis methods can be used to identify variables and create process patterns, such as cross-case analysis, and grounded theory. All patterns are integrated into the developed PRM. They are then represented in a standard form using BPMN. Moreover, a design pattern can be proposed to respond to the current problems. Performance measures are also offered to link the relevant processes leading to operational performance improvement of organisations.

    Stage 3: Application of PRM

    For the application of the developed PRM for healthcare BPR, two approaches should be provided, namely a configurable model and template-based conceptual modelling for simulation. In healthcare systems, there are a large number of processes. Moreover, a variety of business processes have been found. Thus, a configurable model is established to group hospital process patterns together. All variables that affect a sequence of activities will be identified as configurable nodes. Model users can consider these nodes in order to create the specific process model of an organisation for a redesign project. Additionally, it can be of benefit to the model developer for the maintenance of a set of processes.

    Furthermore, changes in healthcare systems can be tested using simulation modelling that allows redesign improvements prior to real-life implementation (AbuKhousa et al., 2014). This approach can lead healthcare stakeholders to participate in a BPR project (Kotiadis et al., 2014). However, simulation modellers spend the most time structuring a conceptual model, which is one of the most difficult tasks in simulation modelling (Robinson, 2015). It is frequently emphasised in the process of developing and using simulation models (Furian et al., 2015). Conceptual modelling is a non-software-specific description of the computer simulation model, describing the objectives, inputs (experimental factors), outputs (responses), content (scope and level of detail), assumptions, and simplifications of the model (Robinson, 2015). These elements can be included into a PRM for facilitating a specification of a simulation (Tolk et al., 2013). A PRM that embeds a more detailed set of guidelines for simulation conceptual modelling can benefit both business users and analysts (Weaver et al., 2012).

    Stage 4: Evaluation of PRM

    At the final stage, a model that is designed and developed based on existing reference artifacts can be generalised through qualitative research methods (Galster and Avgeriou, 2011). The validation of concepts in these cases of PRM is often derived from the generic reference model (Cloutier et al., 2010). Hence, an evaluation stage is less crucial for a PRM that is not built from scratch (Galster and Avgeriou, 2011; Cloutier et al., 2010). However, evaluation is an important task for a PRM construction (Ulrich, 2007).

    As suggested by Galster and Avgeriou (2011), an empirically based PRM can be evaluated by focusing the support on efficient adaptation and installation. This approach can use a participative case study in order to demonstrate the applicability of the developed reference model (Reichert et al., 2013). It is the most favoured way of testing the quality of a PRM empirically in a specific setting. The expected effects can be investigated for leading to acceptance of change in healthcare BPR through an experimental simulation. This keystone corresponds with the purpose of PRM.

    In this paper, the heart of H-PROM has been developed including the process reference framework and hospital reference tasks and a formalisation of BPMN. They are the results of the second stage of the methodological approach that have been constructed based on the reference modelling foundations selected from the first stage. These components are presented in the subsequent section.


    The main aim of the development of PRM in healthcare settings is to provide a standard for facilitating interdepartmental communication and collaboration. To serve this aspect, a specific set of hospital supply chain business processes is firstly framed based on the GSCF framework. Due to GSCF originating in manufacturing, a modification of the framework is necessary in applying the approach to the healthcare industry (Shou, 2013). Each organisation in the supply chain will have its own set of functional silos that must relate to their key SCM processes (Cooper et al., 1997). Therefore, empirical data were collected in hospital case studies for obtaining overall business processes for pharmaceutical product management in hospitals. Based on empirical evidence gathered from these cases and the general theory related to SCM, the process reference framework can be framed by modifying the elements of GSCF. The hospital functional areas and its main tasks as well as the key SCM business processes can also be identified. Additionally, the detailed activities of each task are then analysed amongst the hospital cases. All tasks contained within the framework will also be represented in a standard form using BPMN and exemplified in this section.

    4.1 Pharmaceutical Product Management in Hospitals

    In this study, a multiple case study was considered to gain empirical richness for use in the construction of HPROM. Such studies help to generate generalizable and accurate theoretical insights (Eisenhardt, 1989). Accordingly, the four cases of a large Thai hospital providing tertiary care and a teaching hospital which were convenient to access, were selected for collecting business processes of pharmaceutical product management in hospitals, as follows:

    • Case A was established in 1975 as one of the medical schools in Thailand. It is a 1,100-bed hospital that offers a tertiary care facility in all major fields of healthcare. An average of approximately 2,000 patients per day visit its out-patient department (OPD). This case has more than 800 SKUs (stockkeeping units) of drug items in its inventory.

    • Case B is an 853-bed tertiary care facility that offers specialists in all major fields of healthcare. It was established in 1982 as one of the medical schools in Thailand. This hospital is also recognised as one of the top specialty-referral hospitals in southern Thailand. Today, on average, 3,000 patients visit the hospital daily for OPD service. Hospital B stocks more than 1,350 SKUs in its pharmacy inventory.

    • Case C was established in 1982 as one of the largest university teaching hospitals in Thailand. It offers specialised tertiary care facilities in all major fields of disease with 1,029 beds. At hospital C, on average 4,000 patients per day visit the hospital’s OPD. More than 2,000 SKUs are stocked in its pharmacy inventory.

    • Case D is recognised as one of the first university teaching hospitals in Thailand, and was established in 1888. It provides specialised tertiary care facilities for all major fields of healthcare with a total bed strength of 2,636. At hospital D, on average 10,000 patients per day visit the hospital’s daily OPD service. This hospital stores more than 2,000 SKUs in its pharmacy inventory.

    In these cases, pharmaceutical product management associates with the four main locations including purchasing department, hospital central warehouse, pharmacy rooms, and wards. Each location performs their own operational functions related to drug management including prescribing and dispensing, replenishment, warehousing and distribution, and procurement. Business processes related to these functions are concisely depicted.

    When a patient enters the hospital ward, the demand is generated by the physician for the specific medication. In Thai hospitals, paper-based prescribing is mostly found although a computerized physician order entry (CPOE) system is available. A hand-written prescription or an electronic prescription is then placed at the hospital pharmacy. Afterwards, prescription data are automatically transferred (in case of electronic prescribing) or manually entered (in case of paper-based prescribing) into the information system of the hospital. Each prescription is then prepared for dispensing. After drugs are dispensed, all dispensed items are recorded regarding patient drug treatments and their removal from stock.

    To guarantee supplies in stock, the demand planning tasks are carried out in order to manage hospital inventory in multiple locations. The inventories are controlled by hospital pharmacies and central warehouses. A forecast for determining the reorder point of each drug item is provided. When a stock level decreases to the reorder point, a requisition is made and placed to the hospital warehouse either manually or electronically. A warehouse pharmacist will prepare a replenishment in order to replenish the hospital pharmacy repositories. These replenished drugs are received for storage at the pharmacy’s rooms. They are then available for dispensing.

    To ensure availability of drug supplies, the hospital demand needs to be identified by a procurement pharmacist. The amount of requisitions will be quantified when the hospital inventory on-hand falls below the predetermined stock levels. Purchasing documents are then processed and committed to the suppliers for purchasing. When a shipment arrives, the drug items are inspected by the hospital committee. Drug information is manually entered into the information system of the hospital and then inventory levels are updated. The received drugs are stored under appropriate conditions at the hospital warehouse. These drug items are then ready to replenish the hospital’s pharmacy stock.

    Any unused dispensed drugs that patients do not receive must be returned to the hospital pharmacy in order to be managed properly. The returned items may be brought back to the shelf or sent for disposal. When the drugs are returned, the unused dispensed drugs are checked. Expired drugs are also returned. Then the drug items that need to be returned will be collected and sent to the hospital warehouse. The returned drug items are checked for correctness. A drug return will be coordinated through the purchasing department. The agreement is also assigned with suppliers such as the policy for expired drug returns. Additionally, if a drug recall occurs, the hospital receives recall information and documents that are initiated from the supplier. Then it is forwarded to the hospital pharmacy in order to retrieve the recalled items returned to the supplier. All medication dispensed must be recalled from the patients to the hospital pharmacy. The recalled items in the pharmacy repository will be identified and retrieved to the central warehouse for returning to the supplier. Although a drug recall has never occurred in Thailand, a set of activities is arranged for both hospitals and suppliers, in the event that it ever will.

    Based on the GSCF framework, the empirical phenomena can emerge as an identical structure of hospital supply chain as the core functions of pharmaceutical product management, and the key SCM business processes. Besides, the related tasks that have been implemented inside the intersections of the key SCM business processes and the functional areas can also be recognised. The hospital supply chain process reference framework for pharmaceutical products will be elaborated in the subsequent section.

    4.2 Process Reference Framework

    The provided process reference framework is intended to be comprehended by a business analyst or an organisation. It is composed of two axes, namely hospital functional silos (vertical axis) and the key SCM business processes for pharmaceutical product management that cut across a hospital’s organisation (horizontal axis). Likewise, the operational tasks and its set of activities that reside inside the intersections are also specified as the reference tasks. These elements are defined based on an empirical foundation. They are described as follows.

    4.2.1 Functional Silos (Vertical Axis)

    Pharmaceutical product management in a hospital operates under four main functions. They are prescribing and dispensing, replenishment, warehousing and distribution, and procurement. Consequently, these functional silos can be arranged in a vertical axis as follows.

    The function of prescribing and dispensing of medication is carried out at wards and pharmacy rooms. A doctor prescribes drugs to the patient based on each condition being treated. Then the patient receives their dispensed drugs from a pharmacist. A record of each prescribing and dispensing is also kept. Unused dispensed drugs and recalled drugs are returned to the hospital pharmacy.

    The function of replenishment is performed by a hospital pharmacy in order to operate drug replenishment for each repository. This also involves reverse logistics tasks for managing returned or recalled drugs.

    The function of warehousing and distribution is to store and operate pharmaceutical products before they are distributed to the hospital pharmacy, in order to replenish an internal requisition. The tasks associated with this function are receipt, inspection, storage, picking, packing, and distribution of pharmaceutical products. It also includes the task of drug return management.

    The function of procurement is to guarantee that the hospital has pharmaceutical supplies to meet hospital requisitions. It covers all related tasks of sourcing and order placing. This must also include an agreement between the hospital and their suppliers regarding drug return policies.

    4.2.2 The Key Hospital Supply Chain Management Business Process (Horizontal Axis)

    To gain an efficient and effective SCM, the key SCM business processes should be identified. Accordingly, based on empirical phenomena and the definitions of the eight generic SCM business processes provided by Croxton et al. (2001), the four key SCM business processes can be determined for pharmaceutical product management in a hospital. These processes consist of demand management, order fulfilment, supplier relationship management, and return management. The details will be further elaborated below.

    • Hospital demand management is concerned with balancing the demand that is created by a physician as the drugs needed to treat their patients with the supply chain capabilities. Drug supplies need to be forecast and planned to facilitate fulfilment of supplies on a periodic basis. Demand and sourcing strategy identification are also considered at the time of order processing.

    • Hospital order fulfilment includes all activities necessary to design a network and enable an organisation to meet their customer requisitions (Lambert and Cooper, 2000). Each task in this supply chain process provides timely and accurate delivery for all customers. Requisitions are managed to fulfil stocks in hospital warehouses and replenish an order requested by a pharmacy. Ultimately, there must be sufficient pharmaceutical products and other supplies in hospitals to dispense to patients.

    • Hospital supplier relationship management is a process that represents how an organisation interacts with its suppliers. A relationship between an organisation and its suppliers is managed to reduce cost and enhance efficiency of delivering goods or services to customers (Mettler and Rohner, 2009). Previously, this has been termed the procurement process (Lambert and Cooper, 2000). Then, it was known as supplier relationship management by Croxton et al. (2001) in order to decrease confusion and provide a more comprehensive scope for this SCM process. Thus, a developed purchasing function is included in this process such as Electronic Data Interchange (EDI). In addition, an implementation of material tracking process is also encompassed in order to track and trace the products which are sent to the end customers. Product data from suppliers has therefore to be recorded into all nodes of the supply chain (Guercini and Runfola, 2009). Hence, supplier relationship management in hospital supply chains includes all hospital tasks that associate with their suppliers such as sourcing and order management, a receiving of pharmaceutical products, and supplier data recording throughout the supply chain.

    • Hospital return management is the SCM process by which activities associated with returns, reverse logistics and avoidance are managed within the hospital and across key members of the supply chain.

    4.2.3 Hospital Reference Tasks

    In a hospital, each functional area performs their own operational tasks related to pharmaceutical product management. These tasks should take into account the key SCM business processes (horizon) and coordinate across functional silos (vertical). Based on the GSCF framework, the core phenomena of hospital tasks acquired from hospital cases can be identically explained. Therefore, the core sixteen tasks that reside inside at the intersection of two axes can be defined based on empirical evidence and the general theory related to SCM. All dimensions of the process reference framework for pharmaceutical product management in a hospital are illustrated in Figure 3.

    For each task, a description is provided including its set of activities with the actors. Besides, in some tasks, the authors found that variabilities can be represented by the activities that differ amongst the hospital cases in the manner of implementation for the same task. On the other hand, commonality can be shared amongst these flows. Some activities have been similarly performed in all hospital cases. Hereby, the type ‘commonality (C)’ or ‘variability (V)’ is indicated for each activity. They will be further considered to establish the configurable model. An example of a hospital task with descriptions and related activities defined for H-PROM is presented in Table 1. The complete hospital reference tasks are the subject of Appendix A.

    4.3 Formalisation of BPMN

    In this work, a formalisation of the standardised modelling language is combined as one of components in HPROM. The BPMN notation was selected to represent all hospital reference tasks in the form of a standard business process diagram. This paper only presents the task of ‘drug dispensing’ as an exemplification of a formalisation of BPMN. This task comprises three vertical swim lanes underneath the intersection of the functional division of ‘drug prescribing and drug dispensing’ and the SCM business process of ‘hospital order fulfillment.’ Each lane is an actor of the task namely assistant pharmacists, pharmacists, and patients. The interaction among actors can be displayed through the provided diagram.

    As the defined activities of a drug dispensing task, variability can be found in the activity of ‘Enter a prescription into the system’ that is performed by the assistant pharmacist (c.f. Table 1). The task of drug dispensing starts when a prescription is placed to the hospital pharmacy. After receiving a prescription, the patient is asked to verify their identity. Then a queue is created and the printed queue ticket is given to the patient. In parallel, the patient waits for paying the medication and their prescription is processed based on the queue. For drug dispensing of a hand-written prescription, a list of prescribed items has to be manually entered into the information system for pharmaceutical product management at the hospital. However, this activity cannot occur if a prescription is transferred electronically. Afterwards, the drug labels are printed. If the prescribed drug items are on the shelf, they are picked from the bin based on First Expired, First Out (FEFO) approach and then packed into the labelled bags. A pharmacist checks for correctness. If an incorrect item is found, it will be picked and packed again. Afterwards, prescription information is sent for billing. The packed medications are organised into the queue. Later the patient pays for their expense, and the medications can be dispensed. Prior to dispensing, a pharmacist must identify the right patient and clinically check. If a problem occurs, such as negative drug interactions or allergies, the prescription will be sent back to the physician in order to make a new prescription. The instructions for each drug are explained such as dosage, contraindications, adverse reactions, and storage. The patient receives their medication. However, if the patient would prefer not to receive the dispensed drugs, this activity will be linked to the task of ‘drug return or drug recall.’ Later drugs are dispensed, which will be connected to the task of ‘stock removal and prescription record keeping.’

    As described above, the sequence flow of this activity can be transformed based on the characteristics of drug prescribing tasks. Two ways of prescribing can be found which are paper-based or electronic prescribing. Thus, two business process diagrams emerge in order to process hand-written and electronic prescriptions for drug dispensing. They are illustrated in Figures 4 and 5, respectively.

    Eventually, as shown in Figure 6, all hospital tasks can be standardised using BPMN. These tasks are connected together under the process reference framework. The end-to-end view of a hospital business process for pharmaceutical product management can be provided for each hospital case.


    As it relates to the current situation of Thai hospital supply chains mentioned earlier, BPR is often required to assemble a fragmented piece of process structures for decreasing medication errors and delays in the process. However, unsuccessful projects concerning BPR implementation have been found in healthcare contexts. Moreover, with this system it is difficult to carry out the business process analysis that is the key step of BPR in any context. A process engineer must take a long time to explicitly represent and thoroughly understand the end-toend healthcare process. It is also wasteful if this task must be redone each time. Therefore, our research aims to provide the PRM for supporting a process engineer to achieve success in healthcare BPR. Various practical implications for process engineers and future research are described below.

    5.1 Practical Implications for Process Engineers

    • Establish the competent PRM for BPR in healthcare systems

    This research proposes the structure of a competent PRM for BPR in healthcare systems. It has been formed based on crucial concerns of the context. Accordingly, the three levels of model components should be arranged for providing a holistic approach for facilitating interdepartmental communication and collaboration and acting as an instrument for leading to participation and acceptance of changes. At the top level, the high-level of standard processes’ structure must be initially constructed as a process reference framework. At the next level, a document of the detailed model is combined. This level contains a general description, the related business process diagrams using modelling language techniques, a description of design patterns, and a set of performance indicators. Moreover, other details acquired from the real-life context can be included. Last, an application level is offered in two approaches. First, a configurable model serves as a blueprint for a business analyst or an organisation to create the model in the specific organisation or the project. It can help to maintain the complex set of healthcare business processes. The other one is template-based conceptual modelling for simulations. It should be prepared for modelling a re-design process and leading to acceptance of changes. As described above, the existing reference models cannot support all of these aspects. Our PRM structure is stipulated by composing these components for being an ingredient in healthcare BPR.

    In this article, the heart of the PRM for hospital supply chains of pharmaceutical products has been constructed. They comprise the top-level process framework and the core elements of the detailed model such as a description and a formalisation of BPMN of hospital reference tasks. The developed PRM has been empirically built based on the GSCF framework and the SCOR model. Empirical evidence about the hospital supply chain of pharmaceutical products is included as the key source. Process engineers can reuse and adapt these process knowledge for BPR in this context with reducing timeconsuming and error-prone tasks. Likewise, the PRM can assist all stakeholders to understand an organisation’s relationships through a standard view. Moreover, it could expand the boundary of supply chain operation management towards strategic decision support solutions in this context.

    • Propose an empirically based construction method for developing a PRM for BPR in healthcare systems

    An empirically based construction method for developing a PRM is also introduced. This methodological approach uses empirical evidence as the main source for developing the model. However, a standard procedure has not been found. It can be conducted based on the problem that the PRM is attempting to solve. Thus, according to the structure of PRM for BPR in healthcare systems described earlier, the provided method undergoes four stages including i) an evaluation of reference modelling foundations ii) an empirically based construction of PRM, iii) an application of PRM, and iv) an evaluation of PRM. The outcomes of the first stage, the suitable generic framework and reference model and modelling language technique, will be the foundations for reference process modelling in stages 2 and 3. Eventually, all components can be carried out as pieces of a jigsaw through the related research methods. They will be combined together in order to establish the PRM. Then the applicability of the developed model should be evaluated at the final stage. Process engineers can use this proposed methodology to develop a new PRM for other specified settings in healthcare systems.

    5.2 Implications for Future Research

    Regarding the proposed structure of PRM for BPR in healthcare (c.f. Figure 1), the core components of HPROM have only been empirically developed in this paper. Therefore, future developments are needed to complete the model. The necessary components should be proposed in further research.

    Based on an empirically based construction method, the designed PRM is formed from empirical evidence gathered from hospital case studies. Although the phenomena can emerge from standard hospital tasks, the dissimilarities in the detailed descriptions of a set of activities have been found from the case study data. Besides, a different set of problems that the hospital confronts can be encountered. Accordingly, process patterns should be searched through qualitative analysis methods. Likewise, a description of design patterns or TO-BE should also be offered to overcome hospital supply chain problems.

    However, resistance to change is an obstacle in redesigning healthcare systems. Thus, the developed PRM should be a supportable tool for leading to participation and acceptance of change. To meet this aim, a simulation approach is suggested. Any change can be tested before implementation into real-life contexts. Nevertheless, in simulation modelling, modellers spend most of their time studying the structure of simulation models. It is one of the most difficult tasks (Robinson, 2015), especially in healthcare. Thus, conceptual models should be provided as one application of the PRM. All significant input modelling parameters involved in a simulation study should be determined. A good conceptual model helps a modeller to create the rest of the simulation work straightforwardly. Besides, based on the BPR concept, business users need to take into consideration an improvement in performance measures when a system is redesigned. Performance indicators should also be identified to link with the relevant processes in the model.

    Additionally, healthcare supply chains mostly contain a large collection of processes. Thus, an infrastructure that can identify and keep track of commonalities is required to manage and maintain consistency across variants when the model is updated (Dijkman et al., 2012; La Rosa et al., 2009). Similar process models are combined together in order to illustrate a comprehensive process model which captures the various tasks as a single big picture, and this is called a configurable PRM. Thus, new challenging tasks and opportunities are arising in the field of PRM (Yan and Grefen, 2011; Dijkman et al., 2012). A model user can build their specific business process by configuring a configurable PRM.

    Accomplishing these further developments can establish an effective PRM. However, the developed model should be evaluated by focusing on the support for adaptation and installation in order to ensure empirical validity.


    This project is supported by the Office of the Higher Education Commission and Mahidol University under the National Research Universities Initiative.



    Proposed PRM components for healthcare BPR.


    Empirically based construction method of a PRM.


    Hospital supply chain process reference framework of pharmaceutical products.


    Hospital reference task-drug dispensing of a hand-written prescription.


    Hospital reference task-drug dispensing of an electronic prescription.


    A formalisation of BPMN in all hospital reference tasks.s


    Example of hospital reference tasks-drug dispensing

    Hospital reference tasks
    *The large figure of hospital supply chain process reference framework of pharmaceutical products shown in Figure 3.


    1. E. AbuKhousa , J. Al-Jaroodi , S. Lazarova-Molnar , N. Mohamed (2014) Simulation and modeling efforts to support decision making in healthcare supply chain management., Sci. World J., Vol. ? ? ? ; pp.2014
    2. R. Al-Abri (2007) Managing change in healthcare., Oman Med. J., Vol.22 (3) ; pp.9-10
    3. A. Bask , M. Lipponen , M. Rajahonka , M. Tinnilä (2011) Modularity in logistics services: A business model and process view., Int. J. Serv. Oper. Manag., Vol.10 (4) ; pp.379-399
    4. M. Bertolini , M. Bevilacqua , F.E. Ciarapica , G. Giacchetta (2011) Business process re ??engineering in healthcare management: A case study., Bus. Process. Manag. J., Vol.17 (1) ; pp.42-66
    5. A. Blecken (2010) Supply chain process modelling for humanitarian organizations., Int. J. Phys. Distrib. Logist. Manag., Vol.40 (8-9) ; pp.675-692
    6. L.C.R. Carpinetti , T. Buosi , M.C. GerA3lamo (2003) Quality management and improvement: A framework and a business-process reference model., Bus. Process. Manag. J., Vol.9 (4) ; pp.543-554
    7. M. Chinosi , A. Trombetta (2012) BPMN: An introduction to the standard., Comput. Stand. Interfaces, Vol.34 (1) ; pp.124-134
    8. R. Cloutier , G. Muller , D. Verma , R. Nilchiani , E. Hole , M. Bone (2010) The concept of reference architectures., Syst. Eng., Vol.13 (1) ; pp.14-27
    9. M.C. Cooper , D.M. Lambert , J.D. Pagh (1997) Supply chain management: More than a new name for logistics., Int. J. Logist. Manag., Vol.8 (1) ; pp.1-14
    10. A. Correia , F.B. Abreu (2012) Adding preciseness to BPMN models., Procedia Technology, Vol.5 ; pp.407-417
    11. K.L. Croxton , S.J. GarcA-a-Dastugue , D.M. Lambert , D.S. Rogers (2001) The supply chain management processes., Int. J. Logist. Manag., Vol.12 (2) ; pp.13-36
    12. R. Dijkman , M.L. Rosa , H.A. Reijers (2012) Managing large collections of business process models: Current techniques and challenges., Comput. Ind., Vol.63 (2) ; pp.91-97
    13. K.M. Eisenhardt (1989) Building theories from case study research., Acad. Manage. Rev., Vol.14 (4) ; pp.532-550
    14. I. El Farouk Imane , T. Abdennebi , J. Fouad (2011) Modeling and simulation of hospital supply chain: State of the art and research perspectives, Proceedings of the 4th International Conference on Logistics, ; pp.287-291
    15. S.D. Francis , P.G. Alley (1996) A ?opatient focus review ?? of surgical services., Business Process Re-engineering & Management Journal, Vol.2 (1) ; pp.48-62
    16. N. Furian , M. O’Sullivan , C. Walker , S. Vössner , D. Neubacher (2015) A conceptual modeling framework for discrete event simulation using hierarchical control structures., Simul. Model. Pract. Theory, Vol.56 ; pp.82-96
    17. M. Galster , P. Avgeriou (2011) Empirically-grounded reference architectures: A proposal, Proceedings of the Joint ACM SIGSOFT Conference QoSA and ACM SIGSOFT Symposium, ; pp.153-158
    18. B. Golden (2006) Transforming healthcare organizations, Healthcare Quarterly, Vol.10 (Sp) ; pp.10-19
    19. S. Guercini , A. Runfola (2009) The integration between marketing and purchasing in the traceability process., Ind. Mark. Manage., Vol.38 (8) ; pp.883-891
    20. M. Hammer , J. Champy (1993) Reengineering the corporation: A manifesto for business revolution., Harper Business,
    21. P. Hertz , G. Finke , S. Verhasselt (2011) A process reference model for service delivery in after-sales field service networks, Proceedings of the the International Conference-Advances in Production Management Systems (APMS2011),
    22. R. Iannone , A. Lambiase , S. Miranda , S. Riemma , D. Sarno (2013) Modelling hospital materials management processes., Int. J. Eng. Bus. Manag., Vol.5 (15) ; pp.1-12
    23. W. Kerley , D.C. Wynn , C. Eckert , P.J. Clarkson (2011) Redesigning the design process through interactive simulation: A case study of life-cycle engineering in jet engine conceptual design., Int. J. Serv. Oper. Manag., Vol.10 (1) ; pp.30-51
    24. M. Khabbazi , M. Hasan , R. Sulaiman , A. Shapi ?(tm)i , A. Taei-Zadeh (2013) Business process modelling in production logistics: Complementary use of BPMN and UML., Middle East J. Sci. Res., Vol.15 (4) ; pp.516-529
    25. S. Khodambashi (2013) Business process re-engineering application in healthcare in a relation to health information systems., Procedia Technology, Vol.9 ; pp.949-957
    26. N. Kock , J. Verville , A. Danesh-Pajou , D. DeLuca (2009) Communication flow orientation in business process modeling and its effect on redesign success: Results from a field study., Decis. Support Syst., Vol.46 (2) ; pp.562-575
    27. K. Kotiadis , A.A. Tako , C. Vasilakis (2014) A participative and facilitative conceptual modelling framework for discrete event simulation studies in healthcare., J. Oper. Res. Soc., Vol.65 (2) ; pp.197-213
    28. J. Kriegel , F. Jehle , M. Dieck , P. Mallory (2013) Advanced services in hospital logistics in the German health service sector., Logistics Research, Vol.6 (2-3) ; pp.47-56
    29. D. Kritchanchai , S. Krichanchai (2010) An adoption of vendor managed inventory in Thailand healthcare industry, Proceedings of the 2nd International Conference on Logistics and Transports,
    30. M. La Rosa , M. Dumas , A.H.M. ter Hofstede , J. Cardoso , W. M. P. van der Aalst , A.H.M. ter Hofstede (2009) Handbook of research on business process modeling, Information Science Reference (IGI Global),
    31. M. La Rosa , A.H.M. ter Hoftstede , M. Rosemann , K. Shortland (2008) Bringing process to post production, Proceedings of the International Conference “Creating Value: Between Commerce and Commons”,
    32. D.M. Lambert , M.C. Cooper (2000) Issues in supply chain management: Don ?(tm)t automate, obliterate., Ind. Mark. Manage., Vol.29 (1) ; pp.65-83
    33. D.M. Lambert , S.J. GarcA-a-Dastugue , K.L. Croxton (2005) An evaluation of process-oriented supply chain management frameworks., J. Bus. Logist., Vol.26 (1) ; pp.25-51
    34. M. MacPhee (2007) Strategies and tools for managing change., J. Nurs. Adm., Vol.37 (9) ; pp.405-413
    35. C.R. McConnell (2010) Umiker ?(tm)s management skills for the new health care supervisor., Jones & Bartlett Learning,
    36. T. Mettler , P. Rohner (2009) Supplier relationship management: A case study in the context of health care., J. Theor. Appl. Electron. Commer. Res., Vol.4 (3) ; pp.58-71
    37. A. Molina , J. Garza , G. JimA(c)nez , P. Bernus , M. Fox (2005) Knowledge Sharing in the Integrated Enterprise., ; pp.269-278
    38. H. Nachtmann , E.A. Pohl (2009) The state of healthcare logistics cost and quality improvement opportunities., Plainfield Illinois, KinneyKusek,
    39. P. O ?(tm)Leary , I. Richardson (2012) Process reference model construction: Implementing an evolutionary multi-method research approach., IET Softw., Vol.6 (5) ; pp.423-430
    40. A. Patwardhan , D. Patwardhan (2008) Business process re ??engineering-saviour or just another fad?, Int. J. Health Care Qual. Assur., Vol.21 (3) ; pp.289-296
    41. S. Ponis (2013) Modeling supply chain processes: A review and critical evaluation of available reference models, Proceedings of the 2nd International Symposium and 24th National Conference on Operational Research, ; pp.270-276
    42. M. Rabe , F.W. Jaekel , H. Weinaug (2006) Reference models for supply chain design and configuration., Proceedings of the 38th Winter Simulation Conference, ; pp.1143-1150
    43. A. Rebuge , D.R. Ferreira (2012) Business process analysis in healthcare environments: A methodology based on process mining., Inf. Syst., Vol.37 (2) ; pp.99-116
    44. J. Recker , M. Indulska , M. Rosemann , P. Green (2006) How good is BPMN really?, Proceedings of the 14th European Conference on Information System, ; pp.1-12
    45. A. Reichert , B. Otto , H. A-sterle (2013) A reference process model for master data management, Proceedings of the 11th International Conference on Wirtschaftsinformatik, ; pp.817-845
    46. S. Robinson (2015) A tutorial on conceptual modeling for simulation, Proceedings of the 2015 Winter Simulation Conference, ; pp.1820-1834
    47. A-W. Scheer , M. Nüttgens , W. van der Aalst , W. Desel , A. Oberweis (2000) Business Process Management, Springer-Verlag,
    48. D. Schwarting , J. Bitar , Y. Arya , T. Pfeiffer (2011)
    49. Y. Shou (2013) Perspectives on supply chain management in the healthcare industry, Proceedings of the 2nd International Conference on Science and Social Research, ; pp.630-633
    50. E. Stavrulaki , M. Davis (2010) Aligning products with supply chain processes and strategy., Int. J. Logist. Manag., Vol.21 (1) ; pp.127-151
    51. C. Svensson , H-H. Hvolby (2012) Establishing a business process reference model for universities., Procedia Technology, Vol.2012 (5) ; pp.635-642
    52. O. Thomas , B. Hermes , P. Loos , A. ter Hofstede , B. Benatallah , H.-Y. Paik (2008) Business process management workshops, Springer-Verlag,
    53. A. Tolk , Y.S. Diallo , J.J. Padilla , H. Herencia-Zapana (2013) Reference modelling in support of M&S-foundations and applications., J. Simul., Vol.7 (2) ; pp.69-82
    54. F. Ulrich , F. Peter , L. Peter (2007) Reference modeling for business systems analysis., IGI Global, ; pp.118-140
    55. C.N. Verdouw , A.J.M. Beulens , J.H. Trienekens , J. Wolfert (2010) Process modelling in demand-driven supply chains: A reference model for the fruit industry., Comput. Electron. Agric., Vol.73 (2) ; pp.174-187
    56. M. Weaver , P. Albores , D. Love (2012) A Simulation Conceptual Modelling Procedure for SCM Applications, SSRN,
    57. I-L. Wu (2002) A model for implementing BPR based on strategic perspectives: An empirical study., Inf. Manage., Vol.39 (4) ; pp.313-324
    58. Z. Yan , P. Grefen , M. zur Muehlen , J. Su (2011) Business process management workshops, Springer,