Prolonged operation of the gas-transport system in conditions of partial loading involves frequent changes in the volume of gas transportation, which necessitates prompt forecasting of system operation. When forecasting the modes of operation of the gas transport system, the main criterion of optimality implies the maximum volume of gas pumping. After all, in this case, the largest profit of the gas-transport company is achieved under the condition of full provision of consumers with energy. In conditions of incomplete loading of the gas-transport system caused by a shortage of gas supply, optimality criteria change significantly. First, the equipment is operated in ranges far from nominal ones which leads to growth of energy consumption. Secondly, changes in performance cause high-amplitude pressure fluctuations at the outlet of compressor stations. Based on mathematical modeling of nonstationary processes, amplitude and frequency of pressure fluctuations at the outlet of compressor stations which can cause the pipeline overload have been established. To prevent this, it was proposed to reduce initial pressure relative to the maximum one. Calculated dependence was obtained which connects the amplitude of pressure fluctuations with the characteristics of the gas pipeline and the nonstationary process. Reduction in energy consumption for transportation is due to the shutdown of individual compressor stations (CS). Mathematical modeling has made it possible to establish regularities of reduction of productivity of the gas-transport system and duration of the nonstationary process depending on the location of the compressor station on the route. With an increase in the number of shutdown compression stations, the degree of productivity decrease and duration of nonstationarity reduces The established patterns and proposed solutions will improve the reliability of a gas-transport system by preventing pipeline overload and reduce the cost of gas transportation by selecting running numbers of shutdown stations with a known decrease in productivity.