The country-wide power blackout that had struck Pakistan on the night of 9th January may just be the proverbial “tip of the iceberg” and there may be many more waiting to happen. We can fear this because the power grid in Pakistan is currently under tremendous stress that will only grow in the future. Inadequate investment in transmission and distribution (T&D) facilities, a relentless pressure to cut costs from both the government and the regulator, and a continuously weakening of institutional capacity are raising the odds of such incidents. An aggressive and coordinated effort will be needed to prevent such incidents to obviate serious losses to our economy and inconvenience to our people. Power grids, like all human-engineered systems, are prone to failure sooner or later. Even though we can reduce the chances of their failures to some extent through proper design, construction, operation, and maintenance, we cannot eliminate them. At the end of the day, it’s largely a tradeoff between the cost of building additional reliability in the system and its benefits to economy and society. Efforts, therefore, should focus on minimizing the chances of their occurrence in the first place, prevent their spreading into the healthy parts of the grid should they occur, and restore the grid back to normal operation as quickly as practicable to prevent excessive losses and inconvenience. Power grids even in advanced countries occasionally fail as is evident from the widespread blackouts that had struck the United States, Canada, Italy, Sweden, Denmark, Australia, and England during the last decade of the 20th and first decade of the 21stcenturies, collectively interrupting power supply to over 150 million people for many hours. Obviously, developing countries with comparatively weaker power grids are also not immune to such occasional blackout episodes as is evident from the history of power grids in India, Croatia, and Greece. But these episodes there have been only a few and far between. Closer to home, however, and since 2013, we in Pakistan have had a major power blackout almost every year, with 2015 and 2016 even experiencing such blackouts twice in the year. Power grids, like all human-engineered systems, are prone to failure sooner or later. Even though we can reduce the chance of failure to some extent through proper design, construction, operation, and maintenance — we cannot eliminate it entirely Why such failures take place in power grids and why we have not been able to prevent them despite our dazzling advancement in engineering and technology? This has a lot to do with the size and complexity of the grid itself. Large and interconnected power grids enable linking of diverse and distant power generating stations with load centers via complex webs of T&D networks and facilities to benefit from reserve sharing, economic electricity trade, and emergency support, thus enhancing overall system reliability and minimizing supply costs. However, these benefits do not come without their attendant complexity and reliance on a host of supporting services and systems which enable operators to maintain a delicate balance between supply and demand often separated thousands of miles from each other. “Transmission planning” and “system operation and control” are arguably among the most difficult and complex areas in electric power engineering. Transmission planners have to study thousands and thousands of permutations and configurations of the options under their consideration to analyze the impacts of normal, contingency, and emergency conditions on the system in order to ensure its adequacy, security, and stability over the planning horizon. System operators are required to always remain vigilant during real-time operation of the grid to variations in demand and supply and dynamic behavior of various grid components to maintain “frequency” and “voltage” within tolerable limits. Any malfunction or failure of even a minor element in the grid, like a tree branch falling on a distribution line or some careless driver hitting his vehicle with an electricity pole, can trigger a sequence of events that, if left unattended and not isolated, can quickly spread into the whole system leading eventually to its total collapse. This has been proved time and again from the post-incident investigations of the major blackouts some of which we have listed above. A number of efforts are made by grid planners and managers to make their systems and equipment robust and resilient against abnormal conditions that can originate from an unexpected happening outside of their systems such as a sudden increase or decrease of consumer demand, inside of the system like the failure of some key grid component, or acts of nature like inclement weather, storms, or flooding. All of which can, individually or jointly, disturb the delicate balance (in technical terms “synchronism”) of this most complex system devised by humans in the past one-and-a-half century. Multiple tiers of protection, fault handling and clearing equipment, and real-time status monitoring and control systems are deployed in the grid to protect it against a large number of credible and probable contingencies to prevent it from becoming unstable and to get it back to normal with as little human intervention as practicable. Despite all the preventive measures and multiple layers of defense, power grids still collapse and the efforts by analysts in the aftermath of the major blackouts to simulate the collapse by feeding the different expected causes in their system analysis tools have led only to frustration and despair as each episode was found to be unique and the causes attributed to it have been rendered mere conjectures. Realizing the futility of a failure-proof power grid (or its enormous cost), electric utilities and their regulators attempt to mitigate the impacts of such contingencies by devising contingency management plans. Such plans generally aim at segregating a complex grid into potential islands by distributing generation and demand evenly among them, as much as possible, to restrict a contingency to only the affected part and prevent it from spreading into other parts of the grid. Once the fault in the affected portion is cleared, the other islands are re-joined into the original grid. A hallmark of these plans is their field testing at regular intervals by conducting drills to mimic real-life contingencies. Despite all these efforts and preparations, however, an odd blackout escapes these defenses and firewalls. Each such blackout event provides an opportunity, notwithstanding its attendant cost and inconvenience, to critically review the adequacy of our protective systems and safeguards as well as the effectiveness of our plans and preparation to handle such episodes. As such, all such blackout events are invariably followed by formal inquiries to determine the causes of the blackout event and recommend a set of measure to prevent them from recurring in the future. Two formal inquiries were conducted in the aftermath of the January 9th blackout in Pakistan: one by the National Electric Power Regulatory Authority (NEPRA); and the other by the National Transmission and Despatch Company (NTDC). In the second and concluding part of this article, we will discuss the major findings of these two inquiries and the lessons they provide. We will also discuss some others factors which have definitely contributed to this particular blackout and thus must be addressed to prevent the recurrence of similar blackouts in the future. (To be concluded) The writer is a freelance consultant specialising in sustainable energy and power system planning and development