The gas lift, by limited capacity of 25 MMSCF/D, was introduced for Khafji field in 1988 which could successfully sustain target rate until mid of 2004. Even though, additional artificial lift energy was planned as long term solution due to the increase of field water cut and depletion of reservoirs, production optimization was found to be the best way to sustain field target until commissioning of these new facilities. For that purpose, general optimization and gas lift allocation models have been built and applied for Khafji field as presented by Ghoniem et al1.

Although, production system modeling is a powerful technology to optimize and understand the whole production system, the individual well modeling hence dynamic nodal analysis and the reliable field measurements are still highly contributed to increasing oil rate. Consequently, the present work is a further and more comprehensive optimization study for Khafji crude production by:

  • Updating the previously built optimization models

  • Screening for new effective wells for gas lift operation

  • Review the design of existing gas lift installations

  • Optimizing the choke size for many natural flow wells

  • Practical adaptation to the dynamic process of production optimization.

As a result, several high productivity wells were selected and put recently on gas lift operations, also gas lift re-design was made for many wells. Furthermore, wellhead choke size was optimized for many natural flow wells. As a result, a significant marginal oil gain was obtained and confirmed by field measurements. Also, a lot of recommendations for gas lift operations and for mature oil fields were concluded.


Production optimization has been considered as a dull activity in some quarters and has not had the same attention which paid to reservoir simulation which still has some limitations in daily field prediction and optimization. Besides, the impact of surface flow line network is always being considered as fixed input into reservoir simulator, this practice might lead to inaccurate prediction results which are often taken as a design basis for future field development planning.

Therefore, the future technology challenge is either to extent reservoir model to include wells actually up to wellhead then through flow lines until the delivery point, or to link the optimization model with reservoir model as one unit in order to recognize for future uncertainties. However, the continuous updating for production optimization model can overcome these limitations.

Nevertheless, the gain in oil production has to be large enough to warrant the resources necessary for updating the optimization models, because the management has an interest in seeing an increase in oil rate and that production system is operated as efficiently as possible in Khafji field.

Khafji field lies in the Arabian Gulf, 40 km east of Al-Khafji city, Saudi Arabia (Fig. 1). The field is composed of several reservoirs for sweet crude. The most prolific ones are 1st and 2nd Bahrain sand. Where, a strong and an active aquifer provides pressure support and water drive to 2nd Bahrain and to a lesser extent to 1st Bahrain. In addition, four other oil reservoirs are currently active namely Ahmadi limestone, B-limestone, Wara sand and Zubair sand which are commonly referred to as minor reservoirs.

In Khafji field, there are approximately a hundred wells which are already completed with gas lift injection system. But only 44 wells can be operated by the limited capacity of lift gas. Even though, many remaining gas lift wells are still producing naturally with lower rate, some others are currently being shut in. Thus, the most effective wells for gas lift operation were defined by performing lot of screening runs using gas lift allocation models.

This content is only available via PDF.
You do not currently have access to this content.