Computerizing Logistics Management Information Systems

Computerizing Logistics Management Information Systems. Transcription. 1 Computerizing Logistics Management Information Systems A Program Manager s Guide USAID DELIVER PROJECT OCTOBER 2012 This publication was produced for review by the U.S. Agency for International Development. It was prepared by the USAID DELIVER PROJECT, Task Order 4. 3 Computerizing Logistics Management Information Systems A Program Manager s Guide The authors views expressed in this publication do not necessarily reflect the views of the U.S.Agency for International Development or the United States Government. 4 USAID DELIVER PROJECT,Task Order 4 The USAID DELIVER PROJECT, Task Order 4, is funded by the U.S. Agency for International Development (USAID) under contract number GPO-I , order number AID-OAA-TO , beginning September 30, Task Order 4 is implemented by John Snow, Inc., in collaboration with Asociación Benéfica PRISMA; Cargo Management Logistics; Crown Agents USA, Inc.; Eastern and Southern African Management Institute; FHI 360; Futures Institute for Development, LLC; LLamasoft, Inc; The Manoff Group, Inc.; OPS MEND, LLC; PATH; PHD International (a division of the RTT Group); and VillageReach. The project improves essential health commodity supply chains by strengthening logistics management information systems, streamlining distribution systems, identifying financial resources for procurement and supply chain operation, and enhancing forecasting and procurement planning. The project encourages policymakers and donors to support logistics as a critical factor in the overall success of their healthcare mandates. Recommended Citation USAID DELIVER PROJECT, Task Order Computerizing Logistics Management Information Systems: A Program Manager s Guide. Arlington, Va.: USAID DELIVER PROJECT, Task Order 4. Abstract As in-country public health logistics systems become more integrated and sophisticated, many countries are looking to automate their logistics management information systems (LMIS) in order to improve the quantity, quality, and timeliness of logistics data throughout the country. These guidelines were written for managers in the Ministry of Health, program managers, donors, and management information system (MIS) program officers as a reference when considering starting an LMIS automation project, planning for one, and executing that plan. Organized around the phases of a software development project, these guidelines provide practical information and resources to ensure proper program management of a complex technology planning process. Photo credit: Health logistics staff person enters data into a computer. USAID DELIVER PROJECT USAID DELIVER PROJECT John Snow, Inc Fort Myer Drive, 16th Floor Arlington,VA USA Phone: Fax: Internet: deliver.jsi.com. 5 Contents Acronyms Acknowledgments Introduction How to Use This Guide What Is an LMIS? Are You Ready for Automation? Structure of the Guidelines Section One: Getting Started Outline Current Flow of Information and Products Define the Problem Determine If Automation Addresses Problem Map IT Environment, Existing Systems, and Stakeholders Develop a Vision Identify the Project Team Develop a Project Charter Section Two: Planning Facilitate Requirements Gathering Develop Use Cases Design the User Interface and Reports Consider Options for Automation Determine Human and Financial Resources Section Three: Engaging Software Developers Select and Contract Vendors Communicate During the Build Test Software Functionality Perform User Acceptance Testing Develop Training Documentation Section Four: Implementation Develop a Change Management Plan Train Users Roll Out System Section Five: Maintenance Track Bugs Continuously Evaluate Identify Enhancements Plan for Ongoing Technical Support iii v Conclusion 57 References 59 i. 6 Figures Figure 1. Software Development Life Cycle (SDLC) 2 Figure 2.The Integrated Public Health Supply Chain 3 Figure 3. Managing the Development of a Computerized LMIS 6 Figure 4. Sample LMIS Information and Supply Flow Diagram 10 Figure 5. Stages of Automation 13 Figure 6. Organization Chart 20 Figure 7. Sample Wireframe of a Webpage 28 Figure 8.Vendor Evaluation Protocol 39 Tables Table 1. Sample Description of Data Needs and Bottlenecks of Current System 11 Table 2. Project Team Members 19 Table 3. Sample Use Case 27 Table 4.Advantages and Disadvantages of Software Development Methods 32 Table 5. Calculating Total Cost of Ownership (TCO) 33 Table 6.Tools for Managing Software Development 44 Table 7. Deployment Checklist 51 ii. 7 Acronyms CMS COTS CRDM HIS HMIS ICT IT IV&V JAD LMIS MIS MOH RFI RFP SDLC SDP TCO USAID WMS central medical store commercial, off-the-shelf (solution) Collaborative Requirements Development Methodology health information systems health management information system information and communication technology information technology independent verification and validation joint application design logistics management information system management information system Ministry of Health request for information request for proposal software development life cycle service delivery point total cost of ownership U.S. Agency for International Development warehouse management system iii. 9 Acknowledgments The primary authors of this guide Emily Bancroft, Jessica Crawford, Marasi Mwencha, Mimi Whitehouse, Naomi Printz, Joy Kamunyori, and Kate Waldman would like to acknowledge the numerous people involved in the framing, research, drafting, and review of this document, including Ashraf Islam, Ron Pankiewicz, Wendy Bomett-Dodie, Greg Roche, Ashley Smith, and Abdourahmane Diallo. The authors would also like to acknowledge all the USAID DELIVER PROJECT team members who have assisted in the implementation of computerized logistics management information systems during the course of the project. The expertise and lessons learned that resulted from these experiences all contributed to the content of this guide. v. 11 Introduction As countries continue to expand health programs and strengthen the supply chains that support them, there is an increased need for user-friendly tools and software packages to support the timely and accurate collection and reporting of logistics management information. This information can be used for operational decisionmaking, advocacy, and resource mobilization. Automation of a logistics management information system (LMIS) can greatly facilitate the work of supply chain managers by enabling faster collection, transmission, and aggregation of data; by reducing human error in calculations; and by allowing for visibility of data up and down the supply chain. Reducing the time required for data collection, transmission, and aggregation results in data being available more quickly for timely decisions and actions to help ensure products are available where and when needed. Software development and, more specifically, the automation of an LMIS can and should follow project management and information technology (IT) best practices. These guidelines were written for managers in the Ministry of Health, program managers, donors, and MIS program officers as a reference when considering starting an LMIS automation project, planning for one, and executing that plan. By having a common reference to draw on, the authors hope the different managers who are involved in critical decisions of an automation project will understand the steps required to develop a robust IT solution and their role in supporting it. How to Use This Guide These guidelines are intended for program managers who understand how logistics systems operate and who are planning to automate all or part of their LMIS. The creation of an automated, or computerized, system is a process and not an event. The purpose of this guide is to ensure that program managers working on in-country supply chain improvements have the tools and knowledge they need to guide the complex process of overseeing a software design and implementation process. This guide should provide program managers with enough information to understand the process of automating an LMIS; it is not, however, intended to make a program manager an IT specialist or a business analyst. The authors suggest that, whenever possible, the program manager work with a skilled IT technical project manager who has experience in successfully managing the development of computerized information systems. Throughout the guide, the document will refer to various outputs that can be expected during the life of the computerization project. Each time an output is referenced in the text, the icon at left will identify it so the reader is alerted to the purpose of the output and any guidance for its creation. This guide and multiple templates that are referenced throughout the guide are available electronically at The outputs noted in the electronic version will also link to templates or examples of documents that have been created for other projects that program managers can use or reference in their own automation project. 1. 12 Program Manager Responsibilities For the purposes of this guide, program managers are distinguished from other types of managers, such as project managers or technical managers. A program manager might be the manager of a vertical program within the Ministry of Health (MOH) who is responsible for multiple projects, with each project having its own project manager. For an automated LMIS, an IT manager who has technical skills may also provide an important level of management. Throughout the process, committed and skilled staff will be required to design, manage, develop, and use the automated system. However, someone will need to be responsible for pulling everything together, and that responsibility will lie with program manager. In order to ensure rigor is brought to the project management process, the program manager needs to ensure the automation process follows the software development life cycle (SDLC) methodology, a best practice approach toward articulating and constructing software solutions. The program manager should understand the SDLC process so you can use the same vocabulary as your LMIS project manager. The SDLC process guides the development of software systems and the process that people generally use to develop these systems. In any such lifecycle, people, process, and technology all play a role in success. Figure 1 shows one example of the software development lifecycle. Figure 1. Software Development Life Cycle (SDLC) Project Planning Review Project Plan for completeness, feasibility, consistency. Requirements Gathering Examine Requirements for completenss, correctness, testability. Design Analyze Design for coverage, completeness, compliance. Construction Inspect Code for compliance, completeness, correctness. Changes Test Assess Tests for completeness, coverage. Support Deployment Evaluate Quality Status to facilitate deployment decision Ensure Proper Deployment installation, notification, go-live tests. Track Support and Change Management Support-Change-Test-Deployment loop analysis and improvement Source: USAID DELIVER PROJECT, Task Order Turning the Digital Corner. 2. 13 Each donor may also have specific guidelines that you need to follow as part of the software development process, and the program manager is responsible for making sure the project complies with all donor requirements. For example, if you are using U.S. Government funds to support the project, you may be required to show compliance with the SDLC process by providing copies of all documentation to your contract management team or by adhering to specific approval guidelines such as the ADS 548 Independent Verification and Validation (IV&V) Reviews. Although this guide does not go into great detail on specific requirements such as ADS 548, it does provide the framework for compliance with the SDLC process. Figure 2.The Integrated Public Health Supply Chain The program manager will need to manage expectations, generate the appropriate documentation, manage risk, and effectively communicate the status of the project to various key stakeholders. Everyone who comes into contact with an automated LMIS will have expectations about how the system will work, what information it will provide, how the information will be provided, when the system will be completed, and how much it will cost. In today s global economy, implementing partners, donors, and program managers within the MOH are under pressure to provide positive results in a short amount of time and within budget. Managing these expectations requires strong leadership during the implementation of an automated logistics management system. In addition to that, there will be a need, especially in the initiation and acquisition phase, to ensure the requirements (i.e., what the automated system needs to do) are appropriately specified and the timeline for the project and budget are adhered to. What Is an LMIS? An LMIS is a system of records and reports whether paper based or electronic used to aggregate, analyze, validate, and display data (from all levels of the logistics system) that can be used to make logistics decisions and manage the supply chain. A well-functioning LMIS provides decisionmakers throughout a supply chain with accurate, Source: John Snow, Inc Getting Products to People: The JSI Framework for Integrated Supply Chain Management in Public Health. 3. 14 timely, and appropriate data, such as stock on hand, losses and adjustments, consumption, demand, issues, shipment status, and information about the cost of commodities managed in the system. Figure 2 shows a framework for understanding a public health supply chain system. In most systems, the data that describe the logistics system and track movement of products through the system come from periodic reports prepared by the personnel in the facilities who manage the products. Using these data, a computerized LMIS calculates essential stock status indicators and provides information in the form of reports and graphs to decisionmakers to assist with the monitoring of logistics system performance and the planning of future product and logistics system requirements. A computerized LMIS provides several important benefits over a manual LMIS, such as ensuring mathematical accuracy, rapid aggregations, calculations, and productions of reports and graphs. A computerized LMIS also provides functionalities such as alert mechanisms to assist in decisionmaking. An automated LMIS is just one component of the systems that may be in place to manage supply chain operations in your country. There are likely also warehouse management systems (WMS), transportation management systems, as well as billing and accounting systems or any of the myriad other types of systems that support a logistics operation in place already. The difference between an LMIS and the other kinds of systems is the type of decision the system helps to inform. A WMS, for example, helps manage and control products within a warehouse. It also helps manage warehouse processes, such as receiving and storage, order processing picking and packing, and shipping products out of the warehouse. In a public health setting, the collection of data for logistics operations has been, at times, confounded with the collection of data for other purposes, particularly with health information systems (HIS) that track incidence of disease and health service delivery. Logistics information and health information designed to facilitate very different decisions, often by different stakeholders. The types of questions that can be answered using logistics data are: What is the demand for a product through this system over a certain period of time? Does demand change throughout that period? How long will current supplies last? When do we need to procure supplies? Where are our supplies in the pipeline? Do we need to move supplies between tiers in our system? Where is demand the highest? What are the implications for positioning inventory? Are we experiencing losses from the system that require us to take action? Where are losses occurring and what are the causes of these losses? Are supplies flowing smoothly through the pipeline? Do we need to adjust our inventory to account for bottlenecks in the system? Are You Ready for Automation? Before moving forward, it is important to make sure certain factors are in place to ensure the project s success. There are four main success factors that should be in place before moving forward on an automation project. As program manager, you will want to discuss these success factors with stakeholders involved in the process of automating the LMIS, so make sure these critical criteria are in place before you start expending extensive resources on an automation project. 4. 15 Success Factor 1: Strong Existing Logistics Business Processes, or a Commitment to the Time and Resources Needed to Improve Business Processes Before or During Automation While automation enables faster collection, transmission, and aggregation of data, and helps to eliminate human error in calculations, it will not solve every supply chain challenge. Often other interventions are needed prior to automation. As a case in point, automating bad management practices can waste money and make poor processes operate faster while achieving no performance improvement. As another example, if there are data quality issues at one level of the supply chain, automation will not improve the data quality without appropriate efforts in quality assurance, management, supervision, and data collection training. If strong business practices are not present prior to automation, the commitment of time and resources for process improvement will be necessary to ensure that the automation will successfully address the identified supply chain challenges and not exacerbate them. Success Factor 2:A Strong Multidisciplinary Team The implementation of automation initiatives requires thoughtful planning and management to achieve optimal outcomes. Automating any part of a logistics system involves input from a variety of people, including end users, developers, logistics personnel, funders, IT experts, and project managers. No single person will be able to grasp and explain what is needed from every level and dimension of the supply chain. Developing a strong automated system is dependent on having all perspectives involved during the design phase. If you are going to move forward on automating any aspect of your LMIS, you need to make sure that adequate personnel from key areas are available to participate in the design process. In addition, strong and capable leadership is an essential element for overall project success. This leadership should be at all levels, including the project manager, program manager, and executive sponsor levels. This will be discussed in more detail in Section One: Getting Started. Success Factor 3: Long-Term Political and Institutional Support It takes time to develop necessary user requirements for a system, select an appropriate solution, build or adapt solutions to meet the specific requirements, and then implement and roll out the solution. Depending on the solution, the levels at which you want to automate, and complexity of the system, the planning, design, and implementation of an automation project can easily take two to five years. Before moving forward, make sure political and institutional support is available to keep the key stakeholders engaged throughout the whole timeframe. In order to keep momentum during this long process, make sure to have interim milestones and deliverables to implement and test. For example, the automated LMIS could be rolled out in phases over time, starting at the least complex level of the system. Throughout the course of creating a plan, making the process transparent to the MOH and effectively communicating with it will help ensure stakeholders continuous engagement. Success Factor 4:The Resources to Go the Distance Developing an automated system requires significant resources, both initially and throughout the life of the system. This includes not only hardware and software resources, which are required to develop and maintain the LMIS, but also human resources, which are required for sustainability of the system. It is important to consider the anticipated commitment from the individuals working on 5. 16 the system and to establish that adequate protocols are in place to ensure knowledge if key individuals leave the project. The resources needed include both an initial investment for developing and deploying the system as well as continued financial commitment to maintain the system (including future upgrades of software and equipment.) The amount of financial resources required depends on the solution selected. Before starting a project, make sure the total cost of ownership of the system is well understood by those who are supporting the development of the system. Those who are going to be responsible for maintaining it also need to be made aware of the financial commitments. These funds need to be available and committed before you move forward on an automation effort. Securing funding for only the development and hoping that the funds for maintenance, ongoing support, and upgrades will be available in the future will put the project s success at risk. More information about the resources needed for both development and ongoing maintenance of a system is discussed in Section Two: Planning. Structure of the Guidelines These guidelines are structured around the SDLC, which progresses from project initiation to planning, execution, implementation and, finally, closure. Figure 3. Managing the Development of a Computerized LMIS Managing the Process of Developing a Computerized LMIS Phases Outputs Activities Engaging Getting Started Planning Software Implementation Maintenance Developers s /UTLINE CURRENT FLOW s &ACILITATE REQUIREMENTS s 3ELECT AND s $EVELOP A CHANGE s 4RACK BUGS OF INFORMATION AND GATHERING CONTRACT VENDORS MANAGEMENT PLAN s #ONTINUOUSLY EVALUATE s $EVELOP USE CASES s 4RAIN USERS PRODUCTS s $EFINE THE PROBLEM s #OMMUNICATE DURING THE BUILD s )DENTIFY ENHANCEMENTS s 0LAN FOR ONGOING s $ESIGN THE USER s 2OLL OUT SYSTEM s $ETERMINE IF AUTOMA INTERFACE AND REPORTS s 4EST SOFTWARE TION ADDRESSES PROBLEM s #ONSIDER OPTIONS FUNCTIONALITY TECHNICAL SUPPORT s -AP )4 ENVIRONMENT FOR AUTOMATION s 0ERFORM USER EXISTING SYSTEMS AND s $ETERMINE HUMAN AND ACCEPTANCE TESTING STAKEHOLDERS FINANCIAL RESOURCES s $EVELOP TRAINING s $EVELOP A VISION s %STABLISH TIMELINE DOCUMENTATION s )DENTIFY PROJECT TEAM s $EVELOP PROJECT CHARTER s "USINESS CASE s 6ISION STATEMENT s 0ROJECT CHARTER s &UNCTIONAL OR USER REQUIREMENTS s 5SE CASES s,andscape ANALYSIS s 2&02&) s 0ROJECT PLAN s "UDGET s 4EST PLAN s 2&02&) s 323 s 3OFTWARE USER GUIDE s 3ERVICELEVEL AGREEMENT s 3OFTWARE RELEASE PLAN s #HANGE CONTROL PLAN s $EPLOYMENT PLAN s 4RAINING PLAN s 2OLLOUT PLAN s #UTOVER PLAN s "UG TRACKER s,essons LEARNED 6. 17 For the purposes of these guidelines, we renamed the SDLC phases to more meaningful descriptions for a program manager, namely: getting started, planning, engaging software developers, implementation, and maintenance. Each of these phases is made up of a number of key activities, leading to outputs that will be important to the overall management of an automation project. The diagram below shows the phases, activities, and outputs. Descriptions of each phase, the activities necessary to carry out the phase, and templates and guidelines for the outputs are included in each chapter of this guide. The Getting Started section of this guide will walk a program manager through the necessary steps to determine if an automation project is appropriate and feasible in the current environment, inlcuding outlining the current supply chain flow of information and products, identifying bottlenecks and other problems, and mapping the existing IT environment. In addition, this section will guide the program manager through the process of putting together a team and gaining the support of critical stakeholders for the automation project. The section will conclude with the steps needed to create a vision and project charter. Once the project team, vision, and charter are in place, it is time to begin planning for the development of the automated LMIS. This Planning section will guide the program manager through the requirements gathering processes, the development of use cases, and the beginning stages of user interface design. Depending on the size of the project, the program manaager may articulate the project s needs through a consultative process, or coordinate this articulation by making use of IT professionals, typically called business analysts. The program manager will conduct or oversee a landscape analysis of existing solutions and consider the options for automation to determine the best way forward given the stated needs. At that time, more definite numbers can be put toward the resources and timeline, and this section will provide guidance for understanding the total cost of ownership (TCO) of the solution. Just as an architect will draw blueprints before construction begins, the end of this phase is marked by similar blueprints, which are called the system architecture and technical design documents. They will guide the developers who will construct the system. Regardless of the chosen solution, Engaging Software Developers will be necessary to construct or tailor the automated LMIS. Expectations of what IT specialists will do, the titles that are typically used, and the responsibilities they have are covered in this section. This section will guide the program manager through the selection and contracting process for developers. It will also explain the steps of the development process that the program manager will be responsible for, including communication with developers and users, overseeing the functionality testing, and the overall approval of the system. It is critical that the program manager treat this phase as a partnership, not an outsourcing relationship, to ensure that the IT solution will meet the needs of the users. When the automated LMIS has been tested and works to satisfaction, a program manager may ask: Now that I have a system, what next? Implementation begins and introduction of the system to users takes place. In some instances, a pilot helps inform managers what other enhancements are required for the LMIS, and after they are made to the software, a larger roll-out plan is executed. If the automated LMIS is replacing another system (whether paper based or legacy system), a cutover plan will articulate the steps needed to make a smooth transition, including training and deployment 7. 18 options. This section will guide the program manager through the steps toward ensuring a successful implementation of the automated LMIS. Regardless of how well designed the automated LMIS is, users will identify problems and suggest enhancements regarding the performance of the new automated LMIS. Since the resources needed to maintain and support the system are identified and discussed in the Planning phase, the Maintenance section of the guide focuses on the role of the program manager after the system has been implemented and deployed. It is important to continue tracking the success of the system over time, addressing bugs or problems that arise and tracking user needs for future updates or upgrades to the system. A well supported and successful system will be one that is regularly improved based on user experience and evolving stakeholder needs. 8. 19 Managing the Process of Developing a Computerized LMIS Outputs Activities Getting Started Getting Started s /UTLINE CURRENT FLOW OF INFORMATION AND PRODUCTS s $EFINE THE PROBLEM s $ETERMINE IF AUTOMA TION ADDRESSES PROBLEM s -AP )4 ENVIRONMENT EXISTING SYSTEMS AND STAKEHOLDERS s $EVELOP A VISION s )DENTIFY PROJECT TEAM s $EVELOP PROJECT CHARTER s "USINESS CASE s 6ISION STATEMENT s 0ROJECT CHARTER Planning Engaging Software Developers Implementation Section One: Getting Started Maintenance This section will walk the program manager through the necessary steps to determine if an automation project is appropriate and feasible in the current environment, inlcuding outlining the current supply chain flow of information and products, identifying bottlenecks and other problems, and mapping the existing IT environment. Next, this section will guide the program manager through the process of putting together a team and gaining the support of critical stakeholders to ensure the success of the automation project. The section will conclude with the steps needed to create a vision and project charter. Outline Current Flow of Information and Products To help determine which process, decision, or action makes sense to automate, it is important to outline the flow of products and information in the existing logistics system. This can be done by working with the logistics team to develop a simple diagram that shows both the flow of data and the flow of products throughout the logistics system. The purpose of this diagram is to help illustrate the following: The data collected at each level of the health system The forms/tools used to collect these data Points of data aggregation within the system The means by which data move from each level Define the Problem Using the diagram showing the flow of products and data, identify where decisions are made at each level of the system and the data that are needed to inform these decisions. This exercise will help highlight gaps in the information flow or current bottlenecks, where there may be opportunities for improving the system through automation. It may also be helpful to discuss this flow of information with stakeholders to identify their biggest problems in the logistics system. Some of the issues that are often raised include: Lack of data and inability to monitor the functioning of the logistics system at all levels of the health system 9. 20 Figure 4. Sample LMIS Information and Supply Flow Diagram Role MOH Central: Forecast needs Allocate central funds Supervise MOH Central Flow of supplies Flow of data Quarterly Report Medical Stores Central/ Zonal: Procure Store Receive and enter orders Distribute Hospitals : Serve clients Prepare hospital orders and funding Form XB1 (monthly, electronic) Central and Zonal Medical Stores (MS) via MS or hospital vehicle via MS vehicle Form XA2 (monthly, paper) Districts: Review & approve dispensary and health center orders Aggregate data from individual orders into Form XA2 Allocate local funds Deliver to facilities Store supplies in transit Dispensaries & Health Centers: Serve clients Record consumption information Prepare orders Collect local funds Hospitals (Government/FBO/ NGO) via district vehicle Government Dispensaries and Health Centers District Form XA1 (monthly, paper) via district or NGO vehicle NGO Dispensaries and Health Centers CLIENTS Poor quality of logistics data Delays in receiving reports and orders from health facilities Lack of information regarding current stock on hand and resupply quantities needed at service delivery points (SDPs) Delays in shipment of medicines from national level to SDPs Inaccurate forecasts and plans for future investments in medicines and supplies at the national level Limitations in addressing demand-driven need for medicines and supplies, such as seasonal fluctuations of disease burden, outbreaks, or changes in accessibility of SDPs due to weather or geographical barriers. Table 1 shows a sample description of the decisions, data needs, and bottlenecks at various levels of the health system. 10. 21 Table 1. Sample Description of Data Needs and Bottlenecks of Current System Level Decisions Data Needed Level of Data Current Visibility Required Bottlenecks in LMIS Health Facility Medicines and supplies needed for Usage data from previous months Historic data on previous usage (at Orders and consumption data stored next month or next and years facility level) on paper forms; not quarter Population data Previous orders for aggregated for easy HMIS data facility view of historical Budget availability Budget remaining information for facility District Medicine supplies needed for each facility in district Repositioning of stock between facilities to address shortages Forecasting to region/national level for future stock needs Usage data from previous months and years Real time stock information at facility level (if possible) Population data HMIS data Budget availability Historic data on previous usage (at all facilities Previous orders (across district, by facility) Budget remaining (across district, by facility) Orders and consumption data from facilities stored on paper forms; not aggregated for easy view of historical information or for data across the district; currently no real time visibility into stock levels at facility Regional / National Medical Stores Timing of deliveries for new procurements Immediate national Order data from previous months and years Plans of MOH, part Historic data on previous usage (at all facilities) Previous orders at Only order data automated, making other data hard to analyze or use stock needs for ners and donors for all levels for decisionmaking; next six to nine procurement Credit line remain- LMIS not easily conmonths ing for all custom necting with MOH ers procurement records to record upcoming national orders or donor plans for procurement National Level - MOH National stock levels required for three- to fi ve-year forecast Usage data from previous months and years Order data from previous months and years Population data HMIS data/disease incidence survey data Budget availability Plans of partners and donors for procurement Historic data on previous usage (at all facilities) Previous orders at all levels Budget data No access to logistics data stored at medical stores; paper-based records of logistics data at district and regional levels making it difficult to aggregate and analyze in timely fashion; limited access to donor procurement plans to feed overall procurement plan 11. 22 Determine If Automation Addresses Problem Once the problem(s) has been defined, it will be necessary to determine if automation will address it. Automation can facilitate a range of actions, from the collection of data at the facility level, the transmission of these data from the facility to a higher level in the health system, the aggregation of these data at the district, provincial, or central level, to the analysis of the data and the development and transmission of reports back down the system. A few example actions that automation could help with include: Moving data more quickly between levels of the supply chain Calculating accurate resupply quantities for facilities based on previous data Monitoring the quality, completeness, and timeliness of reports and providing that information quickly and accurately Assessing national stock status and providing quick feedback on pipeline issues based on current, past, and future usage patterns Tracking discrepancies of stock (for example, quantities issued vs. quantities received) within the system Investing in any automation effort is a fundamental change to a logistics system, as in most cases it will seek to enhance the performance of the logistics system by introducing new efficiencies in information management and decision- making. As mentioned above, automation will not necessarily solve the logistics system s current challenges. The desired improvements in efficiency and/or effectiveness should justify the cost of an IT investment. Using the list of bot- tlenecks or barriers identified by the team, talk through each one to determine whether automation will help address the identified problem. and the increased efficiency or effectiveness that could be realized by investing in automation. Automation of an LMIS can be very helpful to improving logistics system performance. For example, in Bangladesh, implementation of a web-based LMIS cut the time for LMIS reports from the field to the central level by more than half, from two months to twenty days. However, automation will not help solve all problems in a logistics system. Sometimes investing in a simple solution such as the redesign of a complicated stock card or form to make it simpler for staff to understand and complete accurately can also make a differ- ence in overall system performance. Another intervention, such as a system design, may be necessary before automating any part or level of the existing logistics system. Before embarking on any software development or automation ef fort, one should have in place business processes that support the flow of technically sound infor mation. Every level in the logistics system could benefit from automation, from the service delivery points, intermediary levels, and the central level. However, given the problem(s) that should be addressed, every level may not need automation. Or given the IT environment and the financial and human re sources available, it may not be realistic to try to automate the entire system at the same time. Priori tize the levels that would most benefit from an automation intervention. 12. 23 As shown in the stages of automation diagram below (Figure 5), there are different processes data collection, data storage, data transfer, and data analysis that can be automated, but not all processes need to be done at once. The supply chain in a country comprises several different levels, such as the central level, districts, hubs, and facilities. In some situations, it may make sense to only automate some levels of the supply chain instead of trying to automate the entire supply chain. For example, in a country where there are Internet and electricity challenges in rural areas but reliable Internet and electricity in larger cities, the best option may be to continue to have a paper-based LMIS at the facility level whose data would feed into an automated LMIS at the central and district levels. As mobile technologies become more accessible to people in rural areas, a mobile-based LMIS could be instituted at the facility level to feed into an automated LMIS at higher levels of the system. Data collection, transfer, and analysis could remain paper based, while data storage is automated. Or certain levels of the supply chain could automate while others stayed paper-based. Figure 5. Stages of Automation Stage 1 Stage 2 Stage 3 Stage 4 DATA COLLECTION No electronic collecɵon for logisɵcs data Electronic collecɵon of logisɵcs data minimal if at all LogisƟcs data may be collected through electronic means such as barcoding or scanning LogisƟcs data are likely collected through electronic means such as barcoding or scanning DATA STORAGE Any logisɵcs data captured is collected and stored in paper forms from SDP up through levels of the supply chain Most logisɵcs data collected and stored in paper form at SDP but computrized at aggregaɵon point before the naɵonal level All logisɵcs data computerized from SDP up through levels of the supply chain All logisɵcs data collected and stored electronically between SDP and all levels of the supply chain, providing full access to all logisɵcs data from all points of the supply chain DATA TRANSFER Transfer of logisɵcs data between levels is paper based; computerizaɵon at central level if at all. Limited key data points may be transferred through mobile devices Transfer of data between levels (bidirecɵonal) may be electronic or through printed report Where Internet access is available data transfer may be fully electronic. Where Internet is limited transfer is in digital form (CD, USB drive) but may be physical transfer Data transfer is real Ɵme through a fully networked system, with mulɵple devices in mulɵple locaɵons accessing the same database DATA ANALYSIS Limited to no computerized data analysis available at SDP. Computerized analysis may be available at central level AggregaƟon point (e.g., district, state, zone) and above have access to computerized data for analysis. Limited to no computerized data analysis available at SDP All levels have access to their own computerized data for analysis All levels have access to all computerized data for analysis, limited by user permissions One common pitfall that often happens in software development projects is that the team has a solution in mind before analyzing and understanding the problem. For example, some members of the team may be convinced that a solution using mobile phones or a certain software package is the solution before determining and analyzing the problem at hand. Following the process in this guide should help team members avoid moving too quickly to arrive at a solution before determining the problems to be solved, the specific user requirements for the system, and the infrastructure and technology resources available. Spending time on the planning of the system as a group will lead to better long-term decisions about the technology or the solution needed. 13. 24 Map IT Environment, Existing Systems, and Stakeholders It is important to understand the environment and context in which an automated LMIS will be introduced. Mapping the current IT environment, systems, and stakeholders will help ensure that the automated LMIS will be accepted, sustained, and integrated into current and future IT initiatives in the country. Review and Understand the Existing IT Strategy for the Country Over the last decade, electronic HIS have become a critical part of the monitoring and evaluation and planning strategies of the health sector in most lowincome countries. As a result, ministries of health have been actively building their capacity to plan and manage these investments to make sure they can support and maintain the systems over time. Most ministries of health now have Information and Communication Technology (ICT) departments, which are tasked with doing both long-term planning and ongoing support of electronic HIS. Many countries have written IT strategies specifically for the health sector. Countries often write their ehealth strategies, which are then approved at the national level.a few examples of countries with national ehealth strategies include Kenya,Tanzania, and Ghana. An overview of the current ICT strategy for health in the country and a vision are generally included in ehealth strategies. A more specific road map, guidance on standards and interoperability, costs and benefits, and implementation guidelines also are often included.these documents can help guide a program manager in automation projects. Before getting started with any significant system design, make sure to review and understand the existing IT plans of the health sector. Many of these plans include requirements for coordination with other development efforts, preferences on software development languages and/or platforms, and considerations for support, training, and implementation of any new systems. There may also be requirements about how new systems interact with core existing systems, which may have an impact on the type of system you can implement. Key strategic decisions from the IT plan, such as the need to prioritize the strengthening of existing systems over development or implementation of new systems, could impact the timeline or support for your proposed automation efforts. Understand the Current IT Environment and the Status of Reliable Internet Access, Electricity, Data, and Cellular Networks throughout the Country It is important that the system is designed to work with the infrastructure currently available, with options for growth as infrastructure improves or changes. As described in the introduction, automation can be interpreted in several ways. Automation can happen at different levels of the LMIS for example, only at the central level, where orders are entered, or all the way to the facility level. Depending on the system design, internet access or specialized hardware may be required. If the system design requires facilities to send in order forms via mobile phones, appropriate phones may be necessary, as would strong cellular networks with data capability. Simple SMS systems require just a mobile network. No matter which option is pursued, access to electricity is a key consideration. Once an automated system is in place, users will begin to depend on it, and electricity outages can cause the 14. 25 logistics system itself to break down. Understanding the current IT environment will be essential to the Given the need to collect data in areas successful deployment of an automated LMIS. with unreliable electricity and Internet access, more governments are turning to innovative solutions to ensure the availability Identify Existing Automated Systems of datafrom rural and remote areas. For An LMIS does not operate in a vacuum, but in example, cell phones and tablets are regua dynamic environment, where other automated larly used to collect and send data across systems are already being invested in, planned for, cellular networks, reducing the need for expensive Internet connections at health and implemented. In recognition of these linkages, facilities and hospitals. Partners in mhealth automation of the entire supply chain will involve are working with governments to set up more than just the LMIS. Examples might include low- power computers and servers with a warehouse information system, an enterprise battery or solar backup to maintain crucial resource planning (ERP) system, a transportation software and databases, even in places where power is intermittent or non-exismanagement system, or a vendor management systent.these innovations are expanding the tem. There may be SMS systems already operating possibilities for automation of health data in-country that track stock levels of key tracer com- in many low- and middle-income countries. modities and could therefore integrate with a future automated LMIS. As you start the planning process, one of the first key steps is to understand the other automated systems already in place in the health sector, both within and outside of logistics. The LMIS may also need to interact with other information systems serving other health domains. For example, the routine health management information system (HMIS) may already be automated at a subregional level and could rely on data from the LMIS for key logistics indicators. Or an electronic medical records system could feed data about pharmacy dispensing into an LMIS. Mapping these existing systems is very important during the pre-planning stages. All the stakeholders involved in the automated LMIS process should understand interdependencies that may influence system architecture and design. Any possible areas of overlap with existing systems should be identified and understood before any software development begins. Work with government officials, partners, and health staff to identify the existing systems that operate at the levels of the health system where you will be implementing the LMIS. For each system, try to answer the following questions: What is the purpose of this system? What data does it collect and at what levels of the health system? Does this system have any of the same goals of the proposed automated LMIS? Does it collect any of the same data? Does the automated LMIS need to share any data with this system? Even though you have yet to design your system, answering these questions will help to clearly define the role of the automated LMIS in relation to other existing systems. This will help define the scope of the project and will contribute to the requirements when the team reaches that stage. More information on developing requirements is available in Section Two: Planning. 15. 26 Identify the Key Stakeholder Group for Automated LMIS Design and Implementation Designing and implementing an automated LMIS require input from a broad range of stakeholders. Committed and skilled staff members need to design, manage, develop, and use the automated system for it to be successful. In addition to the staff using the system, other stakeholders should be involved in the automation process. Everyone who comes into contact with an automated system will have expectations about how the system should work, what information it should provide, how the information should be provided, and when the system will be completed. This includes logisticians; medical stores department staff; national government staff involved in clinical planning, procurement, and monitoring and evaluation; ICT staff; donors; and implementing partners involved in supply planning or supply chain strengthening. Engagement with these stakeholders early in the process will help manage their expectations as time goes on. As mentioned previously, maintaining transparency during the automation project will help ensure that stakeholders are engaged and understand what is happening. Because this is an automation process, the initial reaction of stakeholders may be to include in the process primarily technical staff who understand software development. It is very important to have a broad stakeholder group that includes logisticians, planners, managers, and IT staff, especially during the early stages of planning. Levels of knowledge, both about logistics processes and about software or technology design, will vary greatly among the participants. One way to help ensure that all stakeholders regardless of technology knowledge or expertise are able to contribute is to involve a business analyst in the planning process. A business analyst is someone who is trained to understand programmatic needs and processes and translate them into language understood by software developers. Identifying and involving someone with this skill set as a facilitator early in the planning process can improve communication among stakeholders and improve the quality and outcome of the planning process. With all the above information in mind, a program manager should develop a business case document. A business case provides decisionmakers with a tool to analyze and assess options to inform and justify the impact of these project decisions. The business case document generally includes sections on background, current state, objectives, future state, needs, financial analysis, and conclusions and recommendations. Business Case Develop a Vision A clear vision for the automated LMIS, agreed upon by all key stakeholders, is crucial to the success of the automation process. Being able to articulate a shared vision for an LMIS will help stakeholders define requirements for the future system, prioritize needs when resources are limited, and stay focused on the desired impact of the system. In order to create the vision, it is valuable for a program manager to have an executive sponsor and steering committee already organized to assist in the process. The executive sponsor is a manager with demonstrable interest in the outcome of the project. The executive sponsor acts as a vocal and visible champion, legitimizes the project s goals and objectives, 16.