5 new videos are now inside the Classroom; specifically inside online course 120. This course focuses on electricity distribution planning under an uncertain number of electric vehicles (EV) in the grid. The course applies the Stochastic Planning and the Minimax Regret frameworks to this case study. Both frameworks are used in practice for finding the optimal ways to reinforce the electricity grid when there is uncertainty. Each video comes with downloadable Powerpoint slides, Excel files etc. A brief description of each of the 5 new videos follows: - The first video (11 minutes) shows how to calculate the peak electricity demand caused by electric vehicles, in an electricity distribution grid. The calculations, the assumptions, and the structure are identical to what was used in practice. As a reminder, course 120 examines Stochastic Optimization, and Minimax Regret, which are known as "planning frameworks" and are used for finding the optimal investment decisions needed in the grid, when there is uncertainty. - The second video (13 minutes) focuses on the social cost. Specifically , costs are of two types: investment cost and operational cost. The operational cost includes a cost of unserved energy i.e. the welfare loss incurred when electricity demand cannot be served because electricity network lines lack sufficient capacity. Because this cost reflects the welfare loss borne by end-users, we refer to it as a social cost of unserved energy. In this course we focus on the social cost associated with unserved EV charging demand. This cost can be expressed equivalently in £/MWh (per unit of unserved charging electricity) or in £/EV (per vehicle unable to charge). The two units are related through the per-EV charging energy requirement; this video shows how to convert between them. - The third video (11 minutes) shows how to find the number of EVs that cannot be charged with electricity due to network constraints. The social cost of unserved EV charging (in £/MWh) is an input parameter to the optimization. When it is set to a low value, the weight placed on avoiding unserved demand in the objective becomes small relative to the marginal cost of network reinforcement. The optimizer then finds it cost-optimal to defer some line upgrades and accept a positive level of unserved EV charging demand at the optimum; conversely, a high social cost drives earlier and more extensive reinforcement. The video shows for example that in a bus, a total of 0.91 MW of EV peak demand is unserved and it shows how to find that this corresponds to 131 EVs unable to charge.

Was chatting with a portfolio manager on the Exxon stock price. See plot which is here attached. It shows Exxon Mobil’s stock price (XOM) over last year until now. XOM is the ticker symbol for Exxon Mobil on the New York Stock Exchange. On 28 February the US and Israel struck Iran, and because roughly a fifth of global petroleum consumption passes through the Strait of Hormuz, markets instantly priced in a supply disruption and so Brent jumped about 43% in March. XOM tracked it almost tick-for-tick. XOM tracks Brent very closely. See that the stock was relatively flat for much of 2025, then rose sharply in early 2026, and later pulled back a bit. So at some point, the market became much more positive about Exxon, likely because of stronger oil and gas expectations. See for example the second attached plot . The blue line is Exxon Mobil’s stock price, and the orange dashed line is Brent crude oil. Before late February, both move around, but nothing dramatic happens. Then, around Feb 28, both jump sharply: Brent rises from about $70 to about $104 per barrel, and Exxon jumps to a March peak of $176.41. After that, both come down. So the message is that a geopolitical shock pushed oil prices up quickly, and Exxon’s stock moved up with it. When oil goes up fast, investors expect a company like Exxon to earn more cash, so the stock price also rises.

I see many students who want to stay home and so they ask for remote work all the time. Also I see many who go to work and do not dress well. Here is the truth for the job: Truth1: Go 3-4 times every week to the office so others see you. Your presence is very important. Smile and be professional. This plays a role for your promotion. If you stay home most of the time, you will not be promoted and people will forget you. Truth2: Do not use AI a lot at workplace because every computer has trackers and gives your managers a distribution of time (with plots) of how much you used AI and for how long. There is a software (hidden) that summarises your activity on the computer at work. So be careful : if they see that you cannot write code, and all you do is copy-paste from Chat GPT the code, they will fire you sooner or later. Silently one day they will fire you and they will not tell you why most likely. Be careful. Also you can use it on your phone, but it is time consuming.. Ofcourse they have AI tools but you must write code yourself. Truth3: If you go to non-code positions, it is more stressful and more competitive than code positions. Often, you get lower salary also. Because people in energy are scared of coding, and do not like it. So you get an advantage if you can code , understand code etc. Not super . Just basic things. Eg understand Python . Understand ML. etc Truth4: Do not share personal stories with co-workers. They are competing with you for promotion so they want you to fail ... Be careful. Do not share sensitive things. Truth5: Becareful of your social media presence. Managers spy on you e.g. they have fake profiles and are your friends. They will find your second, third etc profiles. They can find your anonymous X profile where you troll people . They have the ability to find you because they have software tools that you do not know . Truth6: yes the online courses here are all you need. But please fix your CV. Align it with energy companies. Do not just design your CV yourself. Get feedback here. Most CVs look rubbish. Try work on your CV carefully.

Just released a new course on energy economics, covering important economic concepts behind investment decisions in electricity distribution networks. You'll learn about - decision frameworks (deterministic, stochastic, least-worst regret), - scenario trees, - stranded assets, - option value of smart technologies, - investment delay. All concepts are illustrated through a practical example. No prerequisites. Ideal if you're preparing for energy economics or power system economics roles, or doing research. It is course 120 at the very end of the Classroom. Briefly here are the definitions of fundamental economic concepts in power systems: - Decision frameworks: these are approaches that network planners use to decide where and when to invest in power systems. These frameworks are: deterministic (ignores uncertainty), stochastic (accounts for uncertainty and probabilities), and least-worst regret (accounts for uncertainty but not probabilities). - Scenario trees: A way to map out possible scenarios. Demand might grow a lot, a little, or not at all. The tree captures these paths and their probabilities. - Investment delay: Some investments take longer to build than others. Upgrading a cable might take years; deploying smart chargers can happen faster. This difference matters hugely for planning. - Stranded assets: You invest in upgrading a line expecting electricity demand to grow, but it doesn't. Now you've paid for capacity nobody uses. That's a stranded asset. - Option value of smart technologies: Smart technologies like smart chargers can be deployed quickly, letting planners wait and see how uncertainty plays out before committing to expensive upgrades. The cost savings from having this flexibility is the option value. - Capitalisation factor: Converts a one-off investment cost into an equivalent annual cost, accounting for the asset's lifetime and the discount rate. Attached is a summary slide, and a slide on the concept of option value and stranded assets. No need to fully understand these screenshots . Just to get an idea of what the course teaches.