TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractCurrent levels of Reservoir Surveillance technology associated with intelligent well completions, such as fibre optics and permanent downhole gauges, create an increasing flow of data.Conventional routine Reservoir Surveillance tools do not help the knowledge worker anymore to cope with highfrequency real-time data. Overloaded with data handling work, the knowledge worker in our industry is not capable to reveal the great potential inherent in this data.A radical different work process and new applications of Data Mining technologies are presented to support the industry's next goal -The Smart Field.A learning Data Mining approach is presented to detect discrepancies from expected trends and patterns. These trends and discrepancies are then translated into business rules to enable the closed-loop control of oil and gas assets.Lessons learned are presented and necessary future developments are identified.

TX 75083-3836, U.S.A., fax 01-972-952-9435. Abstract This paper summarizes the findings of the SPE Forum held in September 2005 on "Making our Mature Fields Smarter".

To assist in the probabilistic forecasting and decision making process for their Captain North Sea heavy oil asset, Chevron has developed an Integrated Asset Model (IAM). This model includes probabilistic predictions of facilities performance and production, enabling decision risk analysis for strategic and operational decisions. The IAM includes risk-based oil, gas and water production forecasts for the Captain Field and cash flows. These forecasts take full account of facilities constraints and uncertainties in reservoir and operational parameters through links to decision risk analysis software. This paper describes the novel approach used and model application. Given the presence of multiple reservoir models, multiple PVT descriptions, three-phase flow, and a variety of well types from infill to ‘new field’, the best source of reservoir performance profiles for each well was the in-house Eclipse™ reservoir simulation models. The production profiles for each well are represented by a rate versus cumulative production curve. A particular feature of the IAM is its ability to automatically capture the output of individual Eclipse™ data sets. The results of the simulation models are combined into a single spreadsheet model. This is of particular importance as it enables the effect of facility throughput limitations on gas and water production to be readily assessed. In particular it provides the ability to choke back individual wells on a priority basis from any of the wells in the separate simulations. A key feature of the model is the ability to investigate new scenarios without the need to run additional Eclipse™ simulations. Validation of this feature is demonstrated. An example of the model's application to a Captain development decision on whether to proceed with a well workover will be discussed. Introduction The Captain Field is a phased development comprising areas A and B (Fig. 1). The field produces heavy high viscosity oil with separation taking place on a Floating Production Storage and Offloading (FPSO) vessel. Critical processing inter-dependencies, ‘single train’ separation with limited redundancy as well as gas and water handling constraints mean that facilities down time can have a significant impact on production projections. There are five producing areas in the field from three reservoir units: Upper Captain Sand (UCS), Lower Captain Sand (LCS) and the Ross. Production consists of the three fluid phases - oil, water, and gas. The gas phase is made up of both free-gas from gas caps and solution gas. Electrical or hydraulic submersible pumps (ESP & HSP) provide the artificial lift in every well. All produced water from the field is re-injected into the reservoir, with a portion of it being used to power the HSPs. The dynamic subsurface behaviour is simulated by four separate reservoir simulation modelsArea A UCS plus the Area B Gas CapArea A LCSArea B Eastern ExtensionRoss Key to the Captain IAM approach is the perhaps surprising result that a wide range of different reservoir simulation, time dependent, production rates from each well can be adequately reconstructed from a two parameter response surface that itself is not a function of time. This response surface is the phase rate tabulated as a function of total down hole reservoir rate and cumulative production.