Higher ed must focus on what AI can make possible and how to best use AI's capabilities with students as the focus on the physical campus.

How AI impacts IT-based connectivity on the phygital campus


Higher ed must focus on what AI can make possible and how to best use AI's capabilities and tools with students as the focus

Key points:

The concept of a “connected campus” has evolved significantly from when it meant accessibility through ease of transportation choices, and the late 1990s and early 2000s emphasis on design for directed flow and gathering spaces in buildings and across physical areas to encourage and catalyze greater social interaction and engagement.

With the increase in the use of IT networks, the concept moved from physical to digital, with a focus on ensuring both integrated architectures for the varied platforms and systems used on a campus and the ability to seamlessly link WiFi devices across campus.

Over the past 5 years, this has evolved even further, merging the physical and digital modalities of a campus into a single integrated “phygital” campus, which results in not just linking functions and functionality in digital space, but, just as importantly, enabling greater connectivity and engagement 24/7, augmenting aspects that are already conducted in physical spaces and traditional modes with those now enabled through technology. The focus is on expanding capability and functionality: Removing the constraints of time and physical co-location, and replacing the frustrating cycle of movement from one office to the next to complete even routine tasks, to one that is seamlessly integrated and user-focused.

Such a campus, representing the merging of digital and physical modalities (resulting in an extension of the physical and time-bound environment through the appropriate incorporation of technology) provides significant opportunities to re-envision higher education, not only making it more accessible to a far larger population but also enabling the transition from teaching (in the sense of a one-way transaction) to learning, with personalization of offerings and time, and support, greatly enlarging the segments of the population served.

“Phygital” campuses combine physical space with integrated digital platforms, enabling institutions of higher education (IHE) to not only reach and serve a larger population but, more effectively, take knowledge to the learner rather than constrain its availability geographically. This shifts the definition and focus of a campus from one built on the concept of primacy of physical space to one that combines the best of physical and digital modalities, enabling an engaged institution where students and knowledge–rather than physical infrastructure–are the focus and are linked through smart networks and platforms, enabling a campus to be defined by continuous and global connections and linkages. While advances in digital technologies and IT are accelerating this transition, it is critical to remember that it is not enough to be “connected.” Rather, the connectivity needs to be effective and efficient, with fast speeds of upload and download, able to handle multiple devices with text, audio, video, and AR/XR without latency, jitter, and packet loss.

The move to immersive modes of learning and remote engagement has created significant, and increased, pressure on connectivity and bandwidth, with new considerations related to an operational environment that is already often severely limited as related to resources and must evolve even with the continued use of legacy and outdated systems that may not effectively “speak to each other.” This takes on additional significance when the knowledge enterprise of an IHE now consists of at least 6 connectivity domains–those of formal learning, research, business operations, holistic student support, social engagement, and a link between academic knowledge and the workplace. This level of interconnectivity can lead to grave unintended consequences when connectivity to one of the constituent platforms/systems gets reduced or cut, emphasizing the overall fragility of such an enterprise.

Modern enterprises operate on data and information, needing rapid and accurate response, with gaps between individual programs and platforms within a system, and across systems, being of extreme concern. Performance degradation as related to loss in data integrity and network speed is a critical challenge, not just within a network, but also across a system comprising multiple networks and many devices. The issue of host identification, which is easily configured manually in small networks, becomes a much bigger issue in these systems as does the issue of configuration conflicts between networks themselves and with myriad devices.

Similarly, the increasing use of massive swarms of data that need to be processed in real time, as well as the continuous integration with IoT devices, results in significant concerns related to speed and capacity, as well as security both within the system and between the system and external devices and networks. The number of layers and types of systems that make up a higher education enterprise network with complex interactions between units across campus, and with vendor systems, adds to the complexity of operation.

It is in this environment that AI tools present tremendous opportunities for academia to enhance security and connectivity without losing the openness, flexibility, and agility that are critically needed to ensure innovation by faculty and students alike. AI provides tremendous potential to enhance their abilities through a combination of four different approaches: (a) Autonomous application of AI where decisions can be made to ensure operational efficiency and continuity, even in light of unforeseen circumstances; (b) Internet-driven application of AI where the response to situations is based on a predetermined set of outcomes based on well-structured specifications, constraints of network and device operations, and overall objectives at the platform, network and systems level; (c) ML-based response, which incorporates inferencing in specified domains both on historical trends and real-time analysis of data on operations and needs; and (d) AI-based continuous optimization that enables greater autonomy to the AI-manager within prespecified domain constraints and objectives.

Faculty and students now use multiple mobile and connected devices, with expectations of seamless connectivity across the physical campus and from remote locations, including international ones. This, combined with the necessity of secure access to online resources over a wide range of applications and services, necessitates dynamic real-time balancing of bandwidth and speed to assure robust and reliable network connectivity within budget constraints. AI tools can assist in this through dynamic balancing between different parts of the phygital campus. The use of ML linked to course schedules, research priorities and deadlines, and data on usage (past and perceived in the future) can be extremely useful. In addition, AI-based platforms can provide tremendous value in areas of recruitment, admissions, and registration beyond assessment of student data and in internal and external communications. An area of growth is related to energy utilization and optimizing facility utilization and occupancy. Intricate, and complex, systems such as finance and compliance that require integration of numerous programs and platforms are at an advantage using AI, which can provide continuing evolution of accuracy even in the face of changing input, constraints, and requirements. While fully autonomous, AI systems for compliance do not exist yet, and perhaps should not–the use of AI to enable constant updating and checking for changes, as well as their application over complex logistical, human, and organizational domains, has significant potential.

Several systems are configured to monitor the use and operational status of devices linked to a network to provide notification of the need for maintenance. With the incorporation of telemetry and AI, monitoring can now be proactive, enabling prognostics rather than just diagnostics. Data on use and stability can be analyzed to predict future events, identify anomalies, eliminate false positives, ensure proactive maintenance, replacement, and upgrades of devices and platforms, and identify and predict future conflicts. This is of special significance when individual programs and platforms on an integrated network are updated or changed, causing inadvertent conflicts that could bring down the entire network. The combination of improved telemetry and AI can also augment overall security through continuous proactive assessment of weaknesses and improved threat mitigation and response.

Similarly, AI-enabled platforms and agents associated with machine reasoning and predictive analytics allow for proactively assessing vulnerability and predicting areas of concern, while also providing multiple action options in real-time. These can be of immense value in assessing network congestion, platform/device conflicts, analyzing flow and peak periods, and enabling complex, multi-protocol integrated responses across the full range of hardware and software operation standards and policies. Multi-agent and knowledge-intensive AI can even suggest changes in protocols for use based on user groups, devices, and application prioritization. The integration of AI into systems at this level can not only enable the identification of areas of operational weakness, but also from a life-cycle perspective, and can be further integrated into dynamic balancing to ward off bottlenecks due to changing needs 24/7 in real-time.

Catalyzed by the COVID pandemic, IHEs are increasingly using hybrid modes of instruction, offering varied levels of traditional face-to-face instruction with digital/online options. This requires two-way streaming and immersive capabilities for 24/7 student engagement and interaction. In addition to the convergence of voice, data, video, and multimedia platforms, there is a need to blend outsourced services with core applications both on-site and in the cloud. These necessitate high levels of stability, sufficient bandwidth, and speeds–often under conditions of limited, rather than open-ended, budgets. This level of integration and optimization benefits from the incorporation of AI at all levels within the network across the enterprise. The open nature of educational campuses results in increased, and unique, considerations of cybersecurity that can address both internal and external users, as well as the integration of outsourced services to seamlessly blend them into a single-sign-on-based enterprise-wide system that enables complex, multi-protocol, and multi-application operations. Manual controls fall short of operational requirements, with AI providing the only effective path forward, especially considering the realities of systems built on the combination of state-of-the-art or near SOTA and legacy systems, as well as that of integrating these legacy systems with cloud and IoT operations within an ecosystem that balances aspects of access, equity, and privacy with those of security.

While tremendous advances have already been made, there are still pain points that are the focus of future development, primary amongst which are the academic realities of an open network with a mix of legacy and new devices operating over complex layers within a disjointed system of central and distributed authority. It will take leadership by IHEs themselves to achieve this, rather than depending entirely on vendors. Similarly, the voice of the full range of IHEs (from community colleges to research universities, and from programs focusing on degrees to those enabling shorter, and more focused, credentials of direct relevance to move up in the workforce) must be included so that local context, including that of resources, is built in ensuring an effective and equitable future for all rather than just at elite institutions which serve a very small minority of all students.

In this vein, higher ed must change its paradigm from fear of AI to one that emphasizes an approach focusing on what can AI make possible and how can we best use its capabilities and tools with students as the focus. Its use in ensuring connectivity and engagement is one such step.

Sign up for our newsletter

Newsletter: Innovations in K12 Education
By submitting your information, you agree to our Terms & Conditions and Privacy Policy.

eSchool Media Contributors

Oops! We could not locate your form.

Sign up for our newsletter

Newsletter: Innovations in K12 Education
By submitting your information, you agree to our Terms & Conditions and Privacy Policy.