The selection of a production strategy is one of the most important tasks to ensure success of petroleum fields development. In order to improve the performance of fields, the use of smart wells is becoming a common practice. In such wells, devices like valves and sensors are able to monitor and control the production, in real time, adding flexibility to the operation. However, it is possible that the expected gain of these wells production does not pay off the required additional investments. Thus, the aim of this research is to compare the differences between smart and conventional well performances.

A methodology of production strategy optimization, which considers the availability of different platforms, each one with a particular fluid treatment capacity, was developed and applied to both the conventional and smart wells. Special care was given to some details of the methodology in order to guarantee a fair comparison between the two options. The methodology was applied to a heterogeneous reservoir, considering a deterministic case. After it, the sensibility of the model was studied by changing geological characteristics, adding uncertainties to the problem.

The optimization results showed small differences between the two alternatives. Consequently, smart wells were able to improve oil production and reduce water production but the net present value (NPV) indicated that the use of conventional well was, in average, slightly more advantageous. Also, the application of different platforms with different capacities induced changes in the performance of the two kinds of wells. Finally, a sensitivity analysis and a detailed discussion were made in order to study situations where the smart wells could be applied.

With the developed methodology, it was possible to carry out a fair comparison between the smart and conventional wells, providing good confidence and efficiency in the optimization process.


One of the major objectives in a petroleum reservoir study is to predict production based on geologic, fluid dynamics, technical and economics parameters. However, the practice to predict the performance of fluids production using reservoir dynamic modeling is tightly associated to the techniques of uncertainty analysis and risk mitigation. The uncertainties and, consequently, the risk will always be present on a field life production due to the insufficient quantity of information to build the model, especially geological data, and on account of the future uncertainties on the oil price.

Generally, these uncertainties will be reduced during the explotation period, because new information is acquired during production. Alike, emergent technologies as geophysics reservoir monitoring (like permanent gauge, 4D seismic) and optimum field management (smart wells, intelligent fields) contributes expressively on the dynamics modeling and consequently on the uncertainties mitigation.

Smart wells are able to improve the field production. These wells have valves, those can be controlled independently, and sensors, those are capable to measure parameters like fluid's composition and well's pressure and temperature. For this reason, smart wells can be employed in projects involving different objectives like control production of gas and water, production by different zones in stratified reservoirs, margin fields, among others.

However, those wells require significant additional investments. Initials studies have been shown that, in some cases, the expected gain of these wells production does not pay off their additional investments. So, it is important to emphasize the need of the study considering different cases, so that it will be possible to know the possible advantages and disadvantages of this relatively new technology. In a study of a new field, specifically considering the production definition stage with the use of reservoir simulation, the valve control strategy can be studied in two different manners: proactive and reactive control.

This content is only available via PDF.
You can access this article if you purchase or spend a download.