Strategic Planning for Rapid and Massive Digital Transformations – An eFactory Model

I teach two graduate level courses regularly at Harrisburg University: ISEM500 (Strategic Planning for Digital Transformation) and ISEM540 (Enterprise Architectures & Integration). In my spare time (nights and weekends), I have been working with numerous small to medium businesses and towns (SMBTs) in more than 40 countries (mostly developing and least developed) due to my engagements with the United Nations and more recently with the British Commonwealth.

Due to the gaps in the marketplace for SMBTs globally, I formed a family and friends startup that has done very well especially due to the current focus on digital transformations. However, I have observed the following through my somewhat blurred lenses:

  • Several organizations in the developed countries seem to be pre-occupied with the idea of digital transformation for improving customer retention and efficiencies in their supply chains. This is understandable, but we are living in an age where the demand for rapid digital transformations is extremely high to support the basic necessities in life such as health, education, public safety, public welfare, food and agriculture, transportation, utilities, housing and many other vital sectors.
  • The economic and human losses due to COVID-19 are staggering and the road to recovery has become longer and more arduous for the poor and unhealthy. To exasperate the problem further, many middle-class families have slipped into the poverty levels (the miles of nice cars waiting for free food over Christmas was a sobering reminder of where we are).
  • Rapid and massive digital transformations are needed globally to vitalize the economies globally. For urgent economic development, new initiatives such as Blue Economies and Manufacturing 4.0 also need significant digital transformations.
  • Unfortunately, many new initiatives are creating point solutions and disconnected “dots” that cause more confusion. According to a small business owner in Michigan –” I run a small business and do not have the time or the background to evaluate thousands of apps that could be of value to me”.

Bottom line: we are facing more serious problems than clickstream mining to determine how many people bought green sweaters last Tuesday! The digital solutions need to address the aforementioned challenges and must be customized for different geographical locations with different limitations. The solutions should be produced quickly and at massive scales to meet the specific demands of the populations. To complicate matters further, several government policies and industry guidelines regulate the deployment and use of some solutions. For example, 3D printing is not allowed by many governments due to potential misuse (e.g., printing firearms). Thus, cookie cutter solutions are not the answer.   

It is virtually impossible to handcraft the needed diverse solutions individually and manually. A factory model is needed to rapidly build highly customized solutions very much like the auto factories that build millions of highly customized cars to satisfy needed safety requirements. For example, safety requirements are not in driver manuals but are embedded in the modern cars. My car beeps at me when I abruptly change the lane – as a reminder of a safety violation.  Modern cars are great examples of solid strategic planning and integrated architectures. These practices need to be replicated widely to address the aforementioned challenges.

Simply stated, a software factory assists in producing software or cyber-physical components (the artifacts) according to end-user requirements through an assembly process. The concept of software factories has been around since the 1960s and has evolved over the years from handwritten process diagrams to the Google Repository in 2020 that maintains all source codes from 25,000 employees of Google [1].  Most software factories so far have been developed for specialized needs of large organizations such as NASA, Airforce, IRS, Google and others. They are not available for general consumption. For example, Google Repository is only available to Google employees [1]. We are interested in a software factory in the cloud that should be smart enough to detect the needs of the intended populations through an intelligent interview, adjust the machinery accordingly to produce customized solutions, and learn for to do a better job in the next round.

We have developed such a factory model [2], introduced in Exhibit1, that quickly produces Smart Collaborating Hubs as shown in Figure 1. These smart hubs contain highly specialized and smart artifacts that are of value to the administrators, end-users, educators, and relevant policy makers on a particular topic. Most importantly, these hubs have prefabricated capabilities for collaboration with each other. Our main requirement is that this factory cannot be a standalone dot – it needs preprocessors and postprocessors to strengthen it. Figure 2 shows a methodology with preprocessors (Phase1) and postprocessors (Phase3) that support the factory operations (Phase2). Needed tools (such as a Digital Transformation Advisor, an ePlanner, and others) are part of the SPACE Factory introduced in Exhibit1. Let us review this methodology by using an example of producing the digital footprint for a telemedicine hub in Jamaica:

  • PHASE1: A user engages in some gamifications to better understand the task and then invokes a Digital Transformation Advisor that suggests the most appropriate services for a telemedicine center in Jamaica. This Advisor guides the user through various scenarios and cost-benefit analysis to determine which technologies available in Jamaica could reduce the costs and increase the benefits of this center. The user can stop here or proceed to PHASE2 to generate the telemedicine hub.
  • PHASE2: The user invokes the ePlanner (the core “factory”) to generate a telemedicine portal plus a strategic plan, feasibility study, funding proposal, an RFP, project management guideline and other artifacts needed to support the hub. To comply with industry practices, the ePlanner also produces an Enterprise Architecture Framework (EAF) based on Togaf (The Open Group Architecture Framework) and a Service Oriented Architecture (SOA) compliant solution architecture.
  • PHASE3: The artifacts generated by the ePlanner are analyzed/revised and then used to generate a final smart telemedicine hub for Jamaica. The final hub is “registered” in a Smart Global Village (SGV), explained later, that interconnects all smart hubs (e.g., the telemedicine center in Jamaica is registered for collaborations with any healthcare facility, smart city, town or community, or educational institution in the SGV network). Any “post-processing” activities (e.g., feeding the ePlanner artifact to a 3D printer) are also conducted in this phase.
  • PHASE4: The results are finalized, and hub administrators go through training for smooth operations. The focus of this phase is to determine Uses Cases that are of high impact but low effort for the users. Specific activities of this phase include detailed implementation scenarios for different user groups, computer aided resource planning for different disaster scenarios, B2B collaboration scenarios between large number of hubs located in different parts of the world, and management and governance issues for workforce training.
A Smart Global Village (SGV)

As a consequence of large number of experiments around the globe, we have generated a Smart Global Village (SGV) — a large sandbox with more than 800 hubs spanning more than 130 countries. Figure 3 displays a conceptual view of the SGV Lab that is being used extensively to develop very interesting and innovative collaboration scenarios for healthcare exchanges, entrepreneurship networks, emarkets and global supply chains. A Global Center provides the subject matter advice and administrative support. We are using a system of systems approach to gradually make the SGV Lab smarter:  a) gradually make the individual artifacts in each hub smarter, b) make each hub smarter by improving the collaboration between portals in each hub, and c) make the overall Lab smarter by making the collaboration between the hubs smarter through better user community involvement. Basically, smart cars gradually become more effective as better digitized road maps become available and smart cars and smart roads, with sensors, learn from each other and become much smarter over time.

Computer Aided Planning, Engineering and Management Methodology based on SPACE Factory

 

I have outlined how a computer aided planning toolkit is evolving into a software factory for digital transformation. We are very encouraged by the results so far and are planning to significantly accelerate our research by continuing our work with the United Nations, the British Commonwealth, and educational institutions.   We have learned that a software factory model, combined with an SGV, is a very valuable tool in teaching graduate courses in strategic planning, enterprise architectures, integration and management of smart cities, communities and enterprises. We are planning to make these tools widely available to other organizations in the future.

 

Key References

 

About the Author: Dr. Amjad Umar is Director & Professor of ISEM Program at Harrisburg University of Science and Technology, an Adjunct Professor of Systems and Telecommunications at University of Pennsylvania, a United Nations Senior Technical Advisor, and a Fulbright Senior Specialist on ICT. He can be reached at aumar@harrisburgu.edu