Greenhouses create a protected environment for the crop,and provide a means for the grower to steer his cultivation.Over the world,the technical options for adjustment of the inside conditions differ widely.The more advanced the greenhouse equipment,the more advanced are the control options,and the more complicated the control task becomes.Classically,local on-off,P and PI controllers were added to each piece of equipment,such as ventilation control,heater control,control of CO2 dosage,fogging control and screens,and more.As a result,the control algorithms have become overcrowded with user defined settings,and the coordination between the loops causes problems.Multivariable control methods,although available in theory,have found little application.In any case,the philosophy is to use control to maintain a pre-set climate.The settings defining the day-night and seasonal trajectories to create acceptable production originate from experience and blue-print advice.The emergence of better crop models and more advanced computing methods allows,in principle,a paradigm shift.Not the indoor climate should be the target,but the cultivation as a whole.Optimal control methodology allows to maximize,on-line,an economic goal function that balances future income from harvesting the crop against current costs of operating the nursery.TheWageningenmethod entails a decomposition of the problem in two parts :an open loop dynamic optimization over the season for nominal smooth weather,plus,on-line,in closed loop,a receding horizon optimization that exploits the short term weather forecast.The result of the dynamic optimization step is a shadow price for a unit of crop biomass,which can then be used as input for the feed-back step to allow the short term balancing of resource costs such as energy and CO2 against the marginal value of the accrued crop.Examples of applications for tomato and sweet pepper show that energy savings of 15% over already optimized classical control can be achieved,under equal production.The temperature and CO2 in the greenhouse are steadily fluctuating;there is no more set-point tracking.Moreover,the optimal control results in much better compliance with hard constraints set by the grower for crop health and vulnerability to pests and diseases.The advantages of this approach are,among others:far less settings by the grower;optimal exploitation of weather;easy adaptation to energy prices.Nevertheless,the method is not easily adopted in the market.Possible reasons might be:the limited scope of models;the feel with the grower to be out of control;perceived less transparent operation;problematic adjustment to equipment change and the level of knowledge required at the computer control suppliers.The talk aims at discussing possible remedies and pathways to promote further economic optimization in greenhouse horticulture operation.