Abstract: Systems and methods for forecasting device return rates are described. Some implementations include initializing a model representing a device's readiness for market based on one or more key performance indicators (KPIs) of the device, the model being represented by a curve, computing differences between measured value of KPIs of the device and values of KPIs fitted to the curve, identifying an inflection point of the curve based on the computed differences, interpreting the shape of the curve based on the identified inflection point, a readiness index and the curvature of the curve and determining a state of readiness of the device based on the interpreted shape of the curve.

Abstract: Systems and methods for evaluating device quality are disclosed.

Abstract: Systems and methods for estimating network capacity or performance are disclosed. In some implementations, a target network resource is selected for evaluation. Key performance indicators for computing a target network resource value for the target network resource are determined. The determined key performance indicators are filtered out of a set of key performance indicators. A data set is gathered. The data set comprises multiple points. Each point indicates a key performance indicator value for at least one of the determined key performance indicators and the target network resource value. A calculation module is trained to calculate the selected target network resource value based on key performance indicator values included in the gathered data set. The calculation module is tested using a testing data set. The testing data set has a same probability distribution of key performance indicator values and target network resource values as the gathered data set.

Abstract: Systems and methods for analyzing and forecasting network traffic are disclosed. In some implementations, multiple points are received. Each point represents a time value and a network traffic value. For each point in the multiple points, the network traffic value is decomposable into a trend component, a seasonality component, a burst component, and a random error component. A trend equation is generated for calculating the trend component based on the time value. A seasonality equation is generated for calculating the seasonality component based on the time value. A burst equation is generated for calculating the burst component based on the time value. A random error equation is generated for calculating the random error component. A network traffic equation is generated for calculating the network traffic value based on the time value by combining the trend equation, the seasonality equation, the burst equation, and the random error equation.

Abstract: Systems and methods for estimating network capacity or performance are disclosed. In some implementations, a target network resource is selected for evaluation. Key performance indicators for computing a target network resource value for the target network resource are determined. The determined key performance indicators are filtered out of a set of key performance indicators. A data set is gathered. The data set comprises multiple points. Each point indicates a key performance indicator value for at least one of the determined key performance indicators and the target network resource value. A calculation module is trained to calculate the selected target network resource value based on key performance indicator values included in the gathered data set. The calculation module is tested using a testing data set. The testing data set has a same probability distribution of key performance indicator values and target network resource values as the gathered data set.

Abstract: Systems and methods for forecasting device return rates are disclosed. Some implementations include identifying correlations between device return cause codes and corresponding device return rates for one or more devices, computing, based on the identified correlations, correlation indexes for each of the device return cause codes, ranking the computed correlation indexes associated with each of the device return cause codes, selecting, a subset of device return cause codes, determining a relationship between the selected subset of the return cause codes and corresponding device return rates, mapping one or more measured key performance indicators (KPIs) of a particular device to return cause codes of the device, and estimating a device return rate for the particular device based on the mapped KPIs and the determined relationship.

Abstract: Systems and methods for evaluating device quality are disclosed.

Abstract: Systems and methods for evaluating cellular voice call quality are disclosed. In some implementations, training data points are received. Each data point includes a call quality value and values of key performance indicators (KPIs) related to call quality. For each KPI, a linear relationship, having a goodness-of-fit value, between the KPI and the call quality value is determined using the multiple data points. Compulsory KPIs are selected based on the goodness-of-fit values. The training data points are separated into clusters based on the compulsory KPI values and the quality values. For each cluster, a mathematical relationship for calculating the call quality within the cluster is determined based on the one or more compulsory KPIs. A module is generated for predicting the call quality value by combining the determined mathematical relationships.