Predicting the future is a complex and often speculative endeavor, and there is no foolproof method for doing so. You can rely on Astrology and read Nostradamus book "Les Prophéties," a collection of quatrains that he claimed contained predictions about future events. Alternatively, you can check your Tarot cards, runes, tea leaf reading, crystal ball gazing or consult your spiritual medium to see, what the future will bring. I recently came across the fact, that ‘The Simpsons’, a long-running animated TV show, has made several jokes and references that have appeared to predict real-world events or developments. Those include events like Donald Trump’s presidency, Ebola Outbreak, Higgs boson discovery or Lady Gaga's Super Bowl Performance. They also had smartphones years before Apple launched their first Iphone in 2007.
Today I think, it is not too important to know, what is happening in the future, meaning in 2, 5 or 10 years – it is important thou what is likely to happen in the near future and even, if I am getting taken by surprise, it really matters, how I can react really quick.
This is particularly true, if you are a managing a construction project and you need to be very experienced, dealing with all the issues, changes, uncertainties that occur during the execution. With most seniors are about to retire during the coming years, a contractor may have to apply other means to support less experienced, junior project managers to weathering the storms ahead. I guess, we all agree, that machines are able to analyze data a lot faster and more accurate, than humans can. This is why we invented computers and Artificial Intelligence is pretty much on top of everyone’s mind these days. AI analyzes data and derives conclusions or recommendations through a process that typically involves several key steps. Have you ever wondered, how?
Here's a simplified overview of how AI systems work for data analysis and decision-making:
Data Collection: The process begins with collecting relevant data from various sources, such as databases, sensors, social media, or other digital channels. The quality and quantity of the data are crucial to the success of AI analysis.
Data Preprocessing: Raw data is often noisy, incomplete, or unstructured. Data preprocessing involves cleaning, formatting, and transforming the data to make it suitable for analysis. This can include tasks like data cleaning, missing value imputation, and normalization.
Feature Engineering: In this step, data scientists or AI systems select and engineer the relevant features or variables from the dataset. Feature selection and engineering aim to reduce dimensionality and extract the most informative aspects of the data.
Model Selection: AI systems use various machine learning and statistical models to analyze the data. The choice of the model depends on the type of data and the problem at hand. Common models include linear regression, decision trees, neural networks, and more.
Training: To make predictions or recommendations, the selected model needs to be trained on historical data. During training, the model learns the underlying patterns and relationships in the data.
Inference: Once trained, the AI model is used to make predictions or recommendations. It takes input data and applies the learned patterns to generate output, such as classifying objects, making forecasts, or suggesting actions.
Evaluation: It's essential to assess the model's performance to ensure its accuracy and reliability. This is done using various metrics and cross-validation techniques to validate the model's effectiveness.
Feedback Loop: AI systems can continually improve through a feedback loop. User interactions, real-world outcomes, and new data are used to retrain and update the model, ensuring it adapts to changing conditions
Decision-Making: Conclusions or recommendations are derived from the model's outputs. For example, in a business context, AI might recommend marketing strategies, pricing decisions, or product recommendations based on the analyzed data
In many cases, especially in critical decision-making processes, human experts review and validate the AI's recommendations before taking action. This human-AI collaboration ensures safety and ethical considerations are met.
Bear in mind, that AI, to start with, will mainly serve 2 use cases: One being to confirm your project managers gutfeel and second to identify pattern and dependencies, that we may have overseen. This will most likely be phase one in our transition, before we even think about automating certain processes.
We are currently working on industry-specific use cases (focus on project inception, submittals and safety), that involve all of the above steps and I will keep you posted on the outcome. Our program include partner applications from our partner Smartapp. Here are some links, that may help to understand: . Smartapp, Just Ask or Joule or SAP Business AI.
Let us know, what use cases you are working on or have in mind?