#futurist #future
Abstract: Daily transactions on electricity and water software’s can be fed into an AI model, along with log books of complaints that are handled by employees in these departments. Once the software is shared with gen AI, AI can provide its insight, along with other details from grid, power supplies of kinds, and water supplies of types, this information can provide meaningful insight. These can be used to develop poor nations in insightful ways, these can be used in natural calamities such as flood and wild fire times, these can be used in outer planet human settlement once learned. All this to be ready for any situation be it be pandemic or earthquake. The AI can understand complex relationships between customer complaints, supply sources to connect, fault lines, connection times, to even reconnection times.
1. Introduction
Every other sector in world is being automated by AI automation. There are gen AI models that learn form past data and automate the working. The biggest example of this is coding. Coding is being automated by gen AI and the automation is such that now engineers are thinking of their jobs in AI. So, what remains and do we need it, and if we need it what for? Yes, I am talking about AI automation for water and electricity supplies. Electricity grids, water and sewage treatments automations. How can AI automate these things? And how can we people of Earth benefit from it?
Water and electricity are indispensable parts of our lives. We need it daily, but what for AI rationalization is required? Well AI is needed when the supplies are touching boundaries. In the perfect scenarios we all get power and water. But when supplies touch the peak, then we all feel the hit. Yes, when say flood comes, and water and electricity supplies are affected. In such scenarios, how to report errors, how to respond to customers, and how to proceed with supply demands. How much and where to provide supplies to? Where to send how much supplies? Is everyone covered with minimum supplies of water and electricity? We need this automation for “right to equality of services” which we all deserve. When supplies are hit, and when supplies are restored, there is a need of a decision-making algorithm which can predict and allocate right amounts of supplies of electricity and water to the customers.
Decision making by humans can be prone to human errors in pressured times. What if AI assist humans in such scenarios. Not just in flood hit areas, these can be natural calamities hit areas such as earthquake hit area, or cloud burst, or wild fires or extreme heat hit areas as well as snow hit areas. In such situations the humans can make errors such as providing extra supplies to one place, and leaving some people without supplies. Once build these models can provide hope for poor nations, as AI once learned would not cost salary of people, and would rationalize supplies in poor nation for both electricity, and water. Poor nations would become developed with all facilities with this automation.
So how to do it? We do it with AI. AI as an assistant to human workers who are providing these essential services to huge populations of people. So AI helps humans in the following decision making.
-Which supply to connect to which end user cluster
-How to cut powers in which regions to save electricity
-How to provide saved energy/water to untouched areas
-How to respond to complaints so that to cover maximum coverage say in flood hit areas
-How to automatically cut supplies of some region say in extreme heat case
-How to automate electricity and water supplies in villages
-How to supply in poor nations
-How to respond to peoples queries in poor nations from data from developed nations
-How to reply to people in absence of workers/employees say in wild fire zones
Decision making for now is based on experience in working in these sectors. But when need strike, installing the services again takes some time. AI can assist in such and many other scenarios. One major area why this is needed is a pandemic like situations when people were struggling to contain virus from spreading. We all have seen those days; AI can help future for any challenges and AI can assist the processes with humans and when in need without humans too. We need such a system. Such systems can be used in case humans decide to settle on Moon, where there is restricted number of people who can work and AI can take the load. So how to do it? Next section discusses the methodology. The challenges and the dangerous parts are discussed in Section 3 and Section 4 discuss the future works.
2. Methodology
The methodology is based on accessing the hundreds of thousands of lines of log books/software SAAS systems/ databases of electricity and water supplies across nation. This also contain the customer data and the lines purchased. Daily transactions on electricity and water software’s can be fed into an AI model, along with log books of complaints that are handled by employees in these departments. Once the software is shared with gen AI, AI can provide its insight, along with other details from grid, power supplies of kinds, and water supplies of types, this information can provide meaningful insight. Log books contain essential data such as area, usage, problems. The AI can also cluster areas into meaningful regions for better decision making. We won’t be using typical probabilistic or operation research-based models. No, we would be using gen AI to learn from the data and then to predict the next outputs. Figure 1 and Figure 2 explains the process in a crisp way. Models such as latest state of art AI models can be used over for better judgement of queries posed by the customers.
Figure 1: AI Model for Electricity Demand and Supplies
The gen AI understand the log books. The AI understands the relationship between a city and nearest supply center. The AI understand the cluster of data. The AI understand how the human expert on duty to service handled a request. The AI understands how to email to authorities about this. The AI understands how to escalate the process. The AI understands how to switch on supplies of a particular area and manage the demand and supply constraints. The AI optimize the supply as if an optimization algorithm is working.
Figure 2: AI Model for Water Demand, Treatment and Supplies
How does AI do it? Well, the AI reads the log books, SAAS softwares, pictures of plumbing complaints, the log books contains the daily logs of where and what is supplied to a person in a particular city. What is the load and how it was resolved. The logs do not contain just one thing, the complaints, the logs contain all actions, the daily inputs of people who work in the supply companies. The actions taken by the engineers and technicians. This helps AI decide how to take decisions the way humans take decision. So, we have all the data for AI to learn from. So, what are we waiting for?
3. Worst Case Scenarios and Need based Use
There are cases AI can go wrong and hence we need human supervision to check AI decision making. AI is just to assist in case of some event like natural disaster or epidemic. Even if AI helps for few days in flooding it would be a good role model to be installed on Moon or Mars. Yes, AI can go wrong and may cut your electricity. But humans can in a reinforcement way make AI learn. And we note that humans would always use AI to reduce the load on them. So, we’ll use AI for need based use. There must be a button to stop AI when AI starts doing its own imagined workflows.
4. Conclusion and Future Work
As explained in this note, we need AI automation for either the natural disaster times, or epidemic times and these learned models can also be used on proposed Moon/Mars missions to allocate electricity and water in new locations. These models can be used and installed in new cities to be build. Once build these models can provide hope for poor countries, as AI once learned would not cost salary of people, and would rationalize supplies in poor nation for both electricity, and water. Similar things can be worked out for water sewage treatment automations. These things can be automated with AI and help developing countries as well as poor nations, while it provide ease to developed nations. These things need to be installed, learned and processed for a bright future of all of the people of planet.
Artificial Intelligence, Computer Science & Mathematics 