Abstract: Light plant factories (LPFs) can be a promising technology to relieve the food crisis in recent years. However, high energy consumption and costs have been attributed to prolonged lighting operation and the cooling demand. Conventional continuous lighting can also overlap with peak electricity price periods during production, thus intensifying economic pressure. This study aims to propose an energy-saving strategy of an intermittent Light-Emitting Diode (LED) lighting in LPFs. A systematic evaluation of intermittent lighting was also conducted to reduce energy consumption and cost under normal crop growth and quality, according to the peak–valley electricity pricing. An experimental cultivation platform was constructed for an LPF with LED lighting devices. Different lighting modes were implemented using timing switches. Indoor temperature was regulated by a conventional wall-mounted air conditioner, while temperature and humidity sensors were installed to monitor the indoor environment. The electricity consumption of the LED lighting and air-conditioning system was recorded using smart power sockets. Romaine lettuce (Lactuca sativa L. var. romana) was selected as the experimental crop. A continuous lighting treatment (16 h light/8 h dark per day, 16 h light period in total) was adopted as the control group (CK). Three regimes of intermittent lighting were tested: L1 (8 h light/4 h dark/8 h light/4 h dark, 16 h light in total), L2 (10 h light/5 h dark/4 h light/5 h dark, 14 h light in total), and L3 (8 h light/5 h dark/4 h light/7 h dark, 12 h light in total). Growth indicators were measured during harvesting, such as fresh weight, plant height, leaf area, and Soil Plant Analysis Development (SPAD) values. Nitrate content was determined to evaluate crop quality. Indoor air temperature, relative humidity, and electricity consumption were recorded continuously to assess the energy performance and economic benefits of different lighting modes. Results revealed that the L1 regime of intermittent lighting shared no significant differences in the lettuce fresh weight, plant height, leaf area, or SPAD value (P > 0.05), compared with the CK. Meanwhile, the nitrate content in lettuce was reduced by 12.4%, indicating the high nutritional quality under normal plant growth. In terms of energy consumption, the main lighting periods of L1 were arranged to avoid daytime outdoor high-temperature intervals, which effectively weakened the superimposed effect between heat generated by LED lamps and external ambient heat gain. Consequently, the peak power of the air-conditioning system was reduced significantly. Air-conditioning electricity consumption of L1 decreased by 4.7%, and the energy consumption per unit lettuce yield was reduced by 18.6%. In terms of economic benefits, the L1 regime increased the proportion of power consumption during valley- and mid-price periods, whereas lighting duration was reduced during peak-price periods, leading to an 11.7% reduction in LED electricity cost, compared with CK. Furthermore, the total electricity cost per unit lettuce yield under L1 was reduced by about 24.8% using the lighting and air-conditioning systems. In contrast, air-conditioning electricity consumption was reduced by 14.1% and 21.6%, respectively, in the L2 and L3, while the LED power use lowered, due to the shorter lighting duration. By contrast, the intermittent LED lighting achieved time-of-use peak shifting and mitigated superimposed thermal load. Lettuce yield and quality were then maintained to reduce the total energy consumption per unit output under peak–valley electricity pricing. This strategy can provide a reliable, practical, and efficient technical solution for energy conservation and cost reduction in closed artificial light plant factories.