דף הבית » היעד הטכנולוגי

היעד הטכנולוגי

Getting Started with Process Management in Government

The overall goal is that within ten years the Israeli government, both internally and in its citizen facing services, will be the world leader in operational excellence, access and transparency for government processes (formerly called bureaucracy).

This will be achieved through a phased approach towards widespread adoption of process management techniques and technologies in government. In order to be able to get a running start in enabling government process management we’ll need to focus on achieving the greatest benefit to the citizens and the clerks, while keeping (at least initially) modeling, system integration and data integration costs to a minimum. This means that we shouldn’t start with a long IT project first creating the optimal definition of a process, and then implementing using a BPM (Business Process Management) suite, web services and XML (eXtensible Markup Language). The processes should remain unstructured (meaning without a rigorous formal model, even for processes where it is possible to generate such a model once the process is understood). The goal is leverage unstructured process management to start quickly, show a benefit and then migrate the more structured processes to standard BPM systems, probably leveraging on SOA (Services Oriented Architecture) and Web Service methodologies.

The first step is to make sure that we don’t try and define optimal processes, but rather stay aligned to existing real-world processes, while enabling visibility and tracking capabilities. This first step will enable a focus on visibility and accountability, not process optimization. During the implementation the focus should be on capturing participant work products (content) and the managing the handoffs between process participants. Content is the work product generated by a participant in a process, handoffs are the flow of work between participants. Starting this way will help get government agencies comfortable with the visibility and accountability that comes from process management but keep the “technology friction” to a minimum.

As the first phase, the goal is to leverage as much as possible the systems and techniques used as part of existing agency processes. Most of the work consists of filling out  and approving forms (the content), and then moving it to the next step in the process (the handoff or workflow). In this first step, manual hand-offs will be replaced be electronic hand-offs so that the system will have the ability to track and provide visibility into the actual, evolving workflow. Since the hand-offs can take place between agencies, there needs to be a universally acceptable hand-off mechanism. This can be achieved by using email as the universal hand-off mechanism, or providing a web-based hand-off mechanism (i.e. cross agency portal) that every participant can access.

If the forms used exist as part of existing applications (and every process participant has access to that application) then participants should continue to use their existing systems as the mechanism to access and modify the content associated with the process. In this case the handoff is a message notifying the next participant of the need to take action, and a way for them to access the relevant information by reference (e.g. link, access key) or by value (screen-scraping).

If there is no existing application supporting the process (or the application isn’t available to all participants), the agency will need to create an electronic version of the form. This can be done in two ways. The first is to simply create an online editable form that mimics the physical form, and the participants will fill out the electronic form instead of the physical form. The second is to create a schema (XSD) version of the form which can be used to create electronic instances of the form. Each agency will need a way of allowing participant access to these XML forms in a human readable format. In this case, the handoff will also need to provide access to relevant form. If needed, digital signature technology can be leveraged to ensure validity of the form and its data in both methods. Digital signatures can also be used for non repudiation.

In this first phase we assume that a knowledgeable human is required to understand the form, since the system will do only the minimal checks regarding data correctness (those available via XSD). To create a more robust system the schema will need to be augmented to support cross element constraints and other assertions not covered by XSD.

The second method requires more work in the beginning, but will enable the creation of a set of reusable schemas and components for the agencies. The second method also makes it easier to extract and load data from the back-end systems by mapping the XML to queries to the back end databases that store the structured content or by screen-scraping methods for existing applications (in order to support existing business logic), and to share content across different agencies. An agency interested in using the data will need to create a map between the different schemas. These schemas and maps should be made available and shared across agencies.

In this first phase the human participant is the interface between the process steps, process flow and the backend systems – and the process flow is unstructured (it is up to the participant to define the next step). We are assuming a knowledgeable person as part of process (the same as today) – they understand how to enter data into the system, and know the appropriate next step. These flows should be captured in a process “system of record” since as more insight is gathered about how these processes are actually executed, process flows can be analyzed and used to create more structured models (either XPDL or BPEL based) of actual processes. These XML data definitions and BPEL or XPDL flows can be the first step into creating more optimized versions of the existing processes.

The system also needs to contain both personal and management reporting mechanisms. Personal reporting enables each participant to understand where their processes stand, and management reporting provides a birds-eye view of overall process status.