When utility companies sustain damages due to natural or man-made events, they are eligible to receive assistance from members of a mutual aid network, if they belong to one. The decision to help is made based on information from the requester and the assessment of the situation by other members of the assistance network. The objective is to construct a logistic regression model that mimics this decision-making process for a dataset from the Canadian Electricity Association's Annual Working Group. One of the main findings was that the most important factor to predict probability of receiving assistance is distance between requester and potential responder. This result has a number of advantages: finding the distance between members of mutual assistance agreements is easily done; using a physical magnitude to make decisions eliminates subjectivity from the decision making process; distance between members can be used to forecast the behaviour of members of existing assistance networks.