Seasonal load variations in district heating systems are so large that some production units become superfluous during summer operations. There is great economic potential in shutting down these units during summer. The economic benefit of summer shutdown is highly dependent on the timing of the shutdown decision. The optimal shutdown and start-up dates depend on complex weather patterns and vary significantly from year to year. This study introduces three classes of decision rules to help production planners perform economically optimal summer shutdown: a fixed date rule, a heat load based rule and a load based rule augmented with weather forecasts. These decision rules are tested using 38 years of hourly weather data to simulate the heat load in Aarhus, Denmark. The large amount of weather data allows for the creation of highly robust decision rules that account for rare, but costly weather conditions. A fixed date rule allows for planning very far ahead and can reap 90.7% of the potential economic benefit of summer shutdown. A heat load based decision rule can salvage 95.8% of the potential shutdown savings at the cost of shorter planning horizons. Augmenting the load based decision rule with 15 day weather forecasts can boost the performance to 96.5%.