Based on review of overburden collapse mechanism in both coal and evaporites, over the last three decades, the author finds empirical methods that are used for assessing cavability inadequate. These empirical methods suggested by Polish researchers are too simplistic offering limited usefulness for estimating face support loading without quantifying the caving characteristics of the main roof strata, ability to concentrate stress, and release it in preferably nonviolent fashion. With the advances in numerical modeling, it is possible to study strata failure and movements in a jointed rock mass for better assessment of cavability and load transfer using both continuum-based and discontinuous models. However, the challenge remains with having reliable stress measurements in the caved zone and around the excavations for calibrating the computer models and establishing design methodologies in caving mining systems. Direct measurements are needed but are rare because of difficulties of placing load cells in the gob.
Using direct measurements in the gob, back analyses of overburden collapse mechanism, and other measurements over the last three decades in the room-and-pillar mines, the author examines the caving mechanism for western US sedimentary rocks. The significance of joint orientation and persistence is highlighted as well as the critical role of stable designs including the use of barrier pillars strategically located to control load transfer and seismicity in room-and-pillar mines, mining under massive stratigraphic units in some of the Western U.S. mines.
Caving mining systems have been used in the United States and Europe for over a century. However as mining continues toward higher cover, it has become critical to evaluate cave conditions for assessment of load transfer, and seismicity during the feasibility analyses and later in design of mine layouts and support systems. In spite of great advancements in methodologies for estimating insitu strength and deformability of rock masses , the understanding and characterization of caving, load transfer, and seismicity is still at an early stage. Empirical methods commonly used for assessing cavability are inadequate and further affected by our inability to measure load transfer through the gob for different geologic conditions [2, 3, 4]. Direct measurements are rare with the only known available case described in this paper. Empirical methods suggested by Polish researchers  are too simplistic, offering some guidance about ease of caving but lacking any quantification of resulting stress concentration and release of energy in preferably nonviolent fashion.