A win-win digital transformation strategy execution requires an organization to leverage technology to maximize customer experience and business value. A culture and support system that fosters collaboration, innovation, quality products/services, and agility is also a must for effective transformation. Without that the advanced technology by itself will not enable an organization to gain a competitive advantage.

Cloud platforms like Microsoft Azure and AWS include access to advanced technology with artificial intelligence (AI) capabilities. These platforms have cost-effective service offerings. The organization should strategically adopt cloud technology for better customer experience and maximizing business value.

Figure 1: Benefits of Cloud Technology Solutions

On a cloud platform, the organization can create:

• Advanced intelligent ecosystem with AI capabilities that will enable it to understand, and solve customer and business problems, as an organization evolves, and customer needs and technology change
• Platform for experimentation and collaboration to increase innovation rate
• A system to maximize operational efficiency that capitalizes on process automation, prediction and recommendation models, optimization, agility, and innovation
• A system that takes advantage of advanced technology through Infrastructure as a Service (IaaS),  Platform as a Service (PaaS) and Software as a service (SaaS) offerings
• Scalable, and resilient cloud system: the ability to scale up or down and scale out or in based on demand. Opportunity to design more resilient system through redundancy on single or multiple regions, as well as include mechanisms to detect the failure and respond gracefully from failure.
• The system that follows security best practice and is in compliance with regulation and standards. On the cloud platform, you will also have access to tools for monitoring, logging, auditing, detecting, notifying, and responding to security threats.
• The system that takes advantage of different pricing options, and tools that will help to understand usage patterns. Hence the ability to optimize the system for performance and cost.

Advanced intelligent ecosystems with AI capabilities on a cloud platform cost less compare to on-premises supercomputing infrastructure with big data storage and AI capabilities. Cloud platform includes access to various powerful technology tools which are easy to use. Thus allow the organization to focus on important things such as understanding business and customer problems, and making sense of data.

The advanced intelligent ecosystem will enable an organization to respond effectively to challenges related to rapid changes in technology and customer needs, as well as fierce competition. For example, the machine learning models can be used for demand forecasting, predicting customer churn, and the recommendation of a product. It can also help to create a targeted marketing campaign to increase customer acquisition at a low cost. Overall, the meaningful insight will enable the organization to achieve the agility it needs to be flexible and respond to the challenges and opportunities in the most effective way.

The data-driven approach will help to make informed decisions, enhance customer experience and maximize business value. For example, the intelligent ecosystem that takes advantage of the machine learning model will help to extract meaningful insight from data to understand customer problems. By understanding customer problems the organization will be able to improve products or services and develop differentiated products.

Figure 2: Advanced Intelligent Ecosystem

Special care is a must for building an advanced intelligent ecosystem to avoid "garbage in, garbage out". The organization needs to use quality data to train and evaluate predictive and prescriptive models. To continually improve algorithms, collects quality data, make the right observation on data patterns, and ask the right questions. In this process, the organization should use the problems as a guideline to determine the optimal solution.

Additionally, the cloud platforms are suitable for maximizing learning from data, collaboration, monitoring systems and insight gathered by solving problem iteratively.  Experiments will help to validate or invalidate the hypotheses. Hence minimize guesswork and reduce the risk on investment.  

On a cloud platform, you can also ingest data, clean, transform and share relevant data sets for further research with subject matter experts (SME), partners, and other collaborators. Hence the ability to create reliable machine learning models, and understand problems and data patterns. Or you can easily share insight with decision makers and enable them to make informed decisions.

The collaboration and experimentation will help to increase innovation rate as a result of knowledge sharing, and idea exploration and exploitation. The feedback from collaboration, data, monitoring systems and experiment will help to take corrective action. In essence, the organization will have an opportunity to continue to innovate and improve upon the solution iteratively based on feedback. The feedback can also be used to streamline the processes and deliver product/services fast. Altogether they create opportunities to enhance the customer experience, prevent or minimize losses, automate processes, and improve innovation rate, agility, and productivity.

The cloud platforms technology tools are easy to use. For example, I have experimented with Microsoft Azure technology tools including Microsoft Azure Machine Learning and HDInsight.   In addition to other modules, the Azure machine learning studio has modules for a neural network, linear regression, and feature selection. There are also modules to clean data, process, and ability to create custom modules in Python/R. Moreover, it is easy to deploy machine learning model from Azure Machine Learning Studio to even Internet of things (IoT) edge devices. The IoT edge devices make possible to process data from sensors/devices using machine learning model locally.

IoT edge devices modules and other application on a cloud platform can take advantage of Microservices architecture, and containerization. Microsoft Azure and AWS have tools such as Docker container and Kurbenetes orchestrator on their cloud platform. Using Microservices architecture and containerization will lead to modular components, scalable application, and fast iterations, delivery, and feedback.

The IoT edge devices have immense potential to make a SCADA (Supervisory Control and Data Acquisition) system more smart and intelligent with machine learning model. The connection of IoT edge devices on the SCADA system is at the early stage.  The edge devices will reduce latency. You can also aggregate data on edge devices before you send them to the cloud for further analysis. Aggregated data require small storage hence the opportunity to save money.

Although IoT is the buzzword that has gain popularity in the past few years, there is a similarity between IoT and SCADA.  SCADA system has automated processes and processes that require human assistance.  The SCADA system operator control devices and monitor sensors and devices from HMI (human machine interface) remotely. 

Moreover, the SCADA system has an alarm management system and data historians. It also has several proprietary and non-proprietary protocols (Modbus, DH, DH+, DF1, etc.). The network for the SCADA system for some of the industries is isolated from that of the enterprise system to minimize the risk of cyber-attack. Last but not least most SCADA systems don't use advanced analytics to improve its operating efficiency and customer experience cost-effectively.

The advanced analytics will help to uncover hidden information from real-time and historical data and enable the organization to solve even complex problems. The insight from the sensors/devices based on machine learning models can be used to make a prediction related to the potential equipment issues. Even better, in the long run, ability create the reliable model for prescriptive maintenance with an ability to generate alternative solutions to fix the diagnosed problem along with the probability of their success. Some of the benefits of IoT or Industrial Internet of things (IIoT) are:

• Predictive and prescriptive maintenance which will lead to a reduction in maintenance cost.
• Ability to monitor and diagnose devices/sensors in the systems and take predictive action to improve products or protect equipment from damage and avoid costly maintenance.
• Improving operational efficiency: effective and efficient energy utilization, increase productivity, systems automation, decrease downtime, and reduce carbon footprint and operating cost 
• Opportunity to share meaningful insight from data with decision makers that will enable them to make more informed decisions.

The OT (operational technology)/IT (informational technology) convergence, as a result of IoT edge devices on a SCADA system, requires security measures to address the increased risk of vulnerabilities. It requires an understanding of the best security practices, standard, regulation and their implication from IT and OT perspectives. The collaboration of organizations in a similar industry, technology providers, and experts is necessary to broaden understanding of measures to minimize or eliminate the risk of cyber-attack. The organization will also need to formulate policies and procedures to address the security issues.

The strategies for security and governance are necessary for an organization to adopt cloud technologies. Cloud providers (AWS, MS Azure, Google, and IBM) security measures adhere to security best practices. They also have security experts continually assessing and eliminating or minimizing the security threats. These cloud platforms are also in compliance with standards like PCI, HIPAA, SOX, and GDPR. Their security tools include recommendations to solve the non-compliant and security threats identified in a system.

However, security is a shared responsibility between the cloud provider and the other parties involved. The software applications, devices, network, and cloud require proper security measures. For example, developers should follow secure programming standards and practices. It is good practice to validate user inputs, encrypt data, and have in place proper security measures for access control. The configuration for authentication, authorization, network security, and services should be done to minimize or eliminate security vulnerability risks.

The organization should also utilize security tools to monitor, log, audit, notify/alert and detect network threat as well as to apply patches. Moreover, it should use the relevant best practices and standards as a guideline to create security policies and procedures. Since technology and cyber-attacks evolve, the organization should continually evaluate and update its policies and procedures.

The success of cloud transformation will depend on the organization ability to have a clear goal, evaluate the gap, and create an effective strategy. The organization should perform cost, benefits, and risk analysis for moving a particular system to the cloud. Alignment between the transition schedule and an upgrade cycle for the on-premises will lead to more savings.

Cost analysis of 3-5 years period is ideal for comparison of hybrid, cloud, and on-premises. The cloud model has operating expenditure (OPEX). The on-premises or hybrid expenses, on the other hand, comprised of the mixture of capital expenditure (CAPEX) and OPEX.

The transformation can begin with items that have low risk and high return. The feedback from the experiment should be used to refine strategy. For an enterprise, hybrid cloud model might be a better option. It is a good practice to keep sensitive information or legacy systems that are too expensive to move to cloud on-premises. The hybrid model will also help an organization to comply with relevant regulations or legal restrictions.

The organization should allocate time to experiment, transition and finalize migration to the cloud. As well as design, monitor, evaluate and adjust meaningful metrics to measure success. The organization should also make use of relevant design patterns and proven practices for architecting cloud solutions.

Proper cloud governance will allow visibility and control over the resources. The organization should formulate governance policies to manage resources, cost, security, and compliance. It should also take advantage of governance tools to enforce policies and make sure resources created and configured meet business requirements. Furthermore to continually monitor system, evaluate, audit for compliance, and improve the governance without sacrificing agility.

The organization should also use resources utilization feedback to identify inefficiency and optimize cloud resources for cost and performance. For example, the organization can eliminate or resize resources which are underutilized with CPU utilization or over utilized. Configure virtual machine instances to automatically scale in or out.  Turning off the non-production server based on the recurring schedule (nonworking hours) will help to reduce cost.

The overall cost savings will depend on how well the systems are designed and optimized. The cloud model will allow the organization to pay for services it is using, instead of paying upfront on infrastructures that it may not need or may use some of the time. The combination of pricing models will help to reduce the overall cost. The pay-as-you-go is similar to the way you pay for electricity, based on your usage.  However, a commitment of 1 to 3 year to reserve the resources (Virtual Machine instances) will lead to cost savings. The organization can save money by making a long-term commitment for those resources that it knows it will need, and use a pay-as-you-go pricing option for the one they are not sure.

Furthermore, cloud platforms like AWS and Microsoft Azure have tools that make easy to automate configuration, provisioning services, backup and site recovery. From Microsoft Azure or AWS, it is easy to create a testing or development environment when you need it and take it down when you no longer need it. You can also configure continuous integration (CI)/continuous delivery (CD) pipelines with automated/manual gates. The advantage of these is the reduction in time and cost to recover from failure, configure, test, deploy, and support the resources.

Conclusion

The cloud technology can help organizations to enhance the customer experience as well as to improve its operational efficiency, innovation rate, and agility. An organization should take advantage of its unique strength and cloud technology solutions. It should also create favorable conditions, and acquire and develop the necessary resources for successful transformation.

Great innovative products don't come out of thin air they are the result of several experiments, collaborative efforts, and adjustments. Organizations with high innovation embrace experimentation, collaboration, and agility. They understand that even though the failed experiments will be more than home runs, it is the few home runs that will give them a competitive advantage.

The organization should start small and improve the solution iteratively. Remember, it takes time to get things right and to reach a point where you can maximize the value on your good investment. Starting sooner than later will help an organization to develop capabilities it needs to gain a competitive advantage and maintain sustainable growth.