This guide is to help you use and understand **Predictto**, the predictive maintenance tool of **Fracttal**. It's divided into two main parts: The basics of predictive analytics and the different models that you can calculate with **Predictto**; and the second part, we show you how to navigate the page of **Predictto** and use the tools that it has available to you.

**Predictive analysis and Forecasts **

At Predictto we want to help you take the best decisions for youmaintenance management system. For this, we use advanced analyticsto create predictive models and deliver forecasts regarding the future condition of your assets. Our algorithms use Machine Learning in combination with statistical models, to give you a complete forecast,mathematically based, and at the same time simple to understand, of so you can make informed decisions regarding your asset maintenance plans.

**Reliability models **

First, it is necessary to clarify what reliability is: also called probability of survival, reliability estimates the probability of an asset to work without failures, from the beginning of its operation to a certain point in time. Reliability is decreasing in the time, as the asset reaches the end of its useful life.

Reliability models describe the behavior of systems, and are built using mathematical and statistical tools, which allow us to show, explain and predict the ability of these systems to operate without failures over time.

Predictto recommends using reliability models for those assets that are more susceptible to failure, and that do not have continuous monitoring. This is because reliability models use failure records (obtained from the assets' work order history) as training data.

**Degradation models **

In simple terms, degradation models allow estimating how the condition of an asset evolves over time, by analyzing relevant physical and / or chemical variables measured over time. Using Machine Learning and advanced analytics we can predict the values of these variables in the future, and create a forecast of the state of the equipment.

In Predictto, we use an algorithm that allows us to establish a probabilistic model of the behavior of one or more variables over time, using measurements of these variables as training data. With this algorithm it is possible to estimate the behavior of the variables in the future, and thus create forecasts of the condition of the asset over time. These forecasts are associated with a degree of reliability, since they are based on an analysis with mathematical foundations. Therefore, we do not create a blind prediction, but we give you a reliable forecast.

It's important to mention that if you want to carry out this type of analysis you need to have monitoring data, either from meters connected to Fracttal, or from other external meters integrated into your Predictto account.