True integration takes more than mixing and matching oil company disciplines.

Integration probably has been the most fashionable word in the exploration and production domain in recent years. Papers, conferences and technical meetings have focused on the concept and the advantages that can be expected when integration is met.
Professionals belonging to the various reservoir disciplines - geology, geophysics, petrophysics and reservoir engineering - are being taught to work together, to search for some synergy and to integrate their individual pieces of work. Vendors are creating integrated databases, shared-earth models and interoperable applications. Managers are creating asset teams, organizing common working environments and encouraging cross-disciplinary educational courses. There is a general consensus around such initiatives, and no single professional would argue integration is not essential in any project. Integration is one of those magic words that always have a positive meaning however they are applied. "Integrated" is always better than "disintegrated," and it is right to look for some sort of integration.
But what is integration? Since it results in the creation of a more complete or harmonious whole, integration can be considered a process whereby extra value is produced.
This general definition perfectly focuses the problem as the concept applies to any E&P project and particularly to the context of reservoir studies.
The challenge
Integration in reservoir studies is a relatively recent concept. The focus has been toward the implementation of concurrent multidisciplinary work processes that substitute the sequential approach from the geological to the simulation model. The issues posed by this approach to reservoir studies usually are related to the integration of the team members, their working environment, the software applications and the hardware configurations.
However, the constitution of multidisciplinary teams does not in itself result in successful integration and does not guarantee that extra value will be gained. The experience of most companies shows that putting a geologist, a geophysicist and a reservoir engineer in the same working room does not necessarily generate integration. Realizing a true integrated study is a challenge that calls for the identification and the solution of several critical issues.
Integrating the information
Geoscientists and reservoir engineers, unlike other scientists, have limited access to the object of their investigation, the reservoir. The information available to them is also peculiar, for at least three reasons.
First, it is mostly indirect. The only direct access to the reservoir is through coring, but only a limited number of direct measurements, such as porosity, can be realized on core samples. In all other cases, the information has to be derived indirectly, via other types of measurements, which can be correlated to the reservoir parameters of interest through some type of transfer function. For example, in a geophysical survey, travel times are measured and converted into depths through a time-depth relationship. Likewise, during a well test pressures are measured and then converted into a number of parameters of interest like permeability or skin factors, the transfer function being some type of solution of the diffusivity equation.
Second, it is based on a small support volume. With the notable exception of seismic and, to a lesser extent, well testing, all the data geoscientists interpret is relevant to a small or very small support volume that is implicitly assumed to be representative of the whole reservoir. For example, rock wettability is measured on a few 1- or 1 1/2-in. plugs and the conclusions are extended to the whole reservoir. Or diagenetic phenomena are observed through the microscope on a thin section of the reservoir rock and the same phenomena are assumed to have acted throughout the producing unit.
Third, it is varied. Information is gathered in several ways - in the cores, in the borehole or from the surface. The number of methodologies used to infer reservoir properties is surprisingly high, from analog geological outcrop studies to axial tomography on a small plug in the laboratory. To further complicate the issue, the same reservoir property can be computed by means of different methodologies that provide independent estimates at different scale. Figure 1 shows the standard rock porosity model for a shaly sandstone, where it can be appreciated that different tools and measurement techniques are sensitive to different portions of the pore system. The understanding of the characteristics and the limitations of each technique is obviously an essential requisite for a correct calibration of the data and hence the integration of the available information.
When permeability is concerned, the picture is even more complicated, as permeability also depends upon the saturation conditions of the rock. Establishing a representative permeability model for the reservoir under study is, primarily, an integration problem.
Technology allocation
A model is by definition a simplification. The degree of such a simplification or, conversely, the degree of complexity of a given study depends on the information available and the human and technological resources allocated to the project.
An important problem that makes integration difficult is the increasing complexification affecting the individual exploration and production disciplines. Complexification, as defined by N.G. Saleri in "Re-Engineering Simulation: Managing Complexity and Complexification in Reservoir Projects," is "the process of adding incremental levels of detail to a study to represent more rigorously its complexity." To a greater or lesser degree, complexification affects every exploration and production discipline.
New technologies offer the geoscientist powerful tools to investigate the details of a particular problem; however, such detail can be more difficult (and sometimes less relevant) to integrate in the study workflow. In other words, new technologies pose new problems of integration, unknown until a few years ago.
The analysis of a cycle of complexification raises two important points related to the integration process:
• increasing the level of complexity of a particular work does not necessarily ensure improved accuracy in overall results; and
• improved accuracy does not automatically guarantee the compliance with the objectives of the study.
These points are often overlooked. Within each discipline, professionals tend to think that the more detail is used in the analysis, the better the quality of the results. Therefore, the various professionals deliver a product that is the best they can achieve with the available technology, compatible with their professional experience and the time available. In this approach, the best analysis implicitly brings the best results.
A typical example of complexification was seen in a 3-D seismic interpretation that was revised through the analysis of seismic attributes, providing a much more detailed image of the fault pattern. The revised interpretation could delineate faults with throws less then 25 ft (8m). More than 2,000 faults were picked in this interpretation.
However, while the detail attained by this study was remarkable, this same detail could be detrimental in the framework of an integrated reservoir study. Actually, such an interpretation could hardly be maintained in a simulation study with the technology available today. It is impossible to reproduce all these small-scale faults in a normal simulation model, where the practical cell size is bigger than most such features. The most likely reaction of the reservoir engineer would be a simplification of the proposed fault pattern, retaining only the most important faults. In other words, the effort of detailing a fault pattern of the reservoir would be followed by the effort of simplifying the same pattern, the net result being a loss of time.
Therefore, when performing an integrated study, it is important to avoid allocating human and technological resources searching for some false detail or for an accuracy that does not add anything but complexity to the study. The degree of accuracy must always be measured against the overall objective of the study. This is why, where an integrated study is concerned, the concept of "good" or "best" work changes.
The bottom line is that the global objective of the integrated study is the driving issue in terms of project organization. Each discipline must redefine its objectives and tools in order to comply with that global objective. This implies a change of focus in the project organization with respect to the traditional approach.
The role of the project manager
An integrated reservoir study must have a project manager whose responsibility is to successfully achieve the objectives of the study within the allocated budget and timeframe. However, as far as integration is concerned, some specific points can be mentioned where the responsibility of the project manager is direct and very important:
• defining the general objectives of the project and identifying those phases of the work that have an impact on the final results;
• allocating the human and technological resources to the project according to the prioritization of work phases mentioned above;
• guaranteeing the correct integration of the different sources of information;
• making sure the required level of technology is applied within each discipline, avoiding useless complexification; and
• ensuring that factors like lack of communication among team members and low interoperability of the different software applications have no negative impact on project development.
The importance of the project manager can not be overstated. Projects often fail because of bad management, and in many cases the problem can be attributed to a lack of understanding of all the different aspects of an integrated study. Often the project leader may have a strong background in one of the disciplines but know little about the others. For example, when the project manager is a reservoir engineer, the integrity of a sound geological model can be jeopardized to meet history-matching in the reservoir simulation phase. This attitude not only can be dangerous from a technical viewpoint but can also frustrate the geologists who worked hard to produce the best image of the subsurface. Conversely, when the project leader is a geoscientist, the bulk of the work can be allocated to the static part of the study and too little attention paid to the field's production performance and the potentially huge amount of information that can be derived from dynamic data.
Most importantly, the project manager must look for integration within his team. He has to understand all the benefits that could be gained when integration is achieved, but he also has to be aware of the problems that may be encountered trying to integrate people and tools.
As new data and new techniques are available to the geoscientist, the process of understanding the static and dynamic nature of a reservoir is becoming an increasingly complex task. More accurate reservoir models at all scales are at hand, provided that results from individual disciplines are properly integrated.
However, integration is a difficult job that can not be achieved simply through cooperation, results exchange or periodic meetings. In fact, integration can be considered a discipline in itself, with its specific problems, methodologies and solutions, and with an associated professional profile, the project leader.

Comments