Abstract: Influenza is a communicable respiratory illness that can cause serious public health hazards. Due to its huge threat to the community, accurate forecasting of Influenza-like-illness (ILI) can diminish the impact of an influenza season by enabling early public health interventions. Current forecasting models are limited in their performance, particularly when using a longer forecasting window. To be able to extend the forecasting window without a substantial decrease in performance, we propose to use additional features such as weather data. Commonly used methods to forecast ILI, including statistical methods such as ARIMA, limit prediction performance when using additional data sources that might have complex non-linear associations with ILI incidence. We propose a novel time series forecasting method, Randomized Ensembles of Auto-regression chains (Reach). Reach implements random chain ensembles for multi-step time series forecasting using machine learning models. We evaluate this new approach on ILI case counts in Auckland, New Zealand from the years 2015-2018 and compare its performance with other standard methods. The results demonstrate that the proposed method performed better than baseline methods when applied to this multi-variate time series forecasting problem.