Revisiting the Congress Street Open Cut in Chicago
- deepexcavation
- Oct 6
- 5 min read
DeepEX LEM Analysis & Slope Stability
Introduction
The Congress Street “superhighway” project in Chicago, constructed in 1952, presented engineers with a significant geotechnical challenge. During excavation, the south slope failed over a length of nearly 200 feet, primarily within saturated glacial clay. Figure 1 presents the approximate slip surface, and the one obtained from ϕ = 0 analysis (Skempton & Hutchinson, 1969). The event was carefully documented and analysed by Ireland (1954), becoming one of the most cited case studies in slope stability.
The case is revisited this time using the Limit Equilibrium Method (LEM) with the Morgenstern-Price (1965) approach in DeepEX, comparing historical analysis with modern software capabilities.
Soil Conditions
Subsurface investigations revealed a sequence of sand and fill materials above glacial clays, divided into three distinct clay layers, with a top sand stratum as presented in detail by Skempton and Hutchinson (1969) (Figure 1):
· +4 to –10 ft: Stiff blue clay with drying cracks, Cu = 1240 lb/ft2.
· –10 to –30 ft: medium strength blue clay, Cu = 820 lb/ft2.
· –30 to –40 ft: stiffer clay, Cu = 1060 lb/ft2.
At greater depth, a stiff clay stratum at around –40 ft controlled the slip surface, while upper brittle clays showed desiccation and progressive failure features, reducing their effective strength.
Figure 1 – Cross-section of the Congress Street open cut with soil profile and failure zone (after Skempton & Hutchinson, 1969).
Failure Observations
Field evidence indicated a nearly vertical escarpment at the crest and a basal crack at the toe of the cut. The observed slip surface was tangential to the stiff clay stratum at depth. Ireland concluded that progressive failure in the upper brittle clay layers was a key factor leading to the collapse.
Figure 2 – Schematic of observed slope failure mechanism (after Ireland, 1954).
Stability Analysis: Ireland vs. DeepEX
Ireland’s original stability assessment was carried out under the φ = 0 assumption, using corrected undrained shear strengths derived from laboratory compression tests. By examining circular slip surfaces, he calculated a factor of safety of approximately 1.1, which indicated that the slope was only marginally stable. This value reflected the precarious condition of the excavation, consistent with the failure that occurred during construction. However, the precise geometry of his computed slip surface did not fully coincide with the observed mechanism. The critical circle extended farther back from the crest than the actual failure surface, a common limitation of analyses restricted to circular geometries.
When the same cross-section was revisited with DeepEX using the Morgenstern-Price method and non-circular slip surfaces, a clearer and more realistic picture emerged. The software predicted a factor of safety of 0.876, which falls below unity and therefore directly indicates instability — in agreement with the fact that the slope did collapse during excavation. Unlike the earlier circular approach, the non-circular analysis produced a slip surface geometry much closer to what was documented in the field. The computed surface initiated just a few feet behind the crest, at an intermediate position between Ireland’s theoretical circle and the actual failure, and curved downward to exit only a short distance from the toe of the excavation.
This closer correspondence between the DeepEX prediction and the observed failure highlights the advantage of modern limit equilibrium methods when freed from the restriction of circular slip surfaces. The non-circular analysis not only provided a more realistic safety margin but also reproduced the kinematics of the actual slide with striking accuracy. Taken together, the comparison illustrates how Ireland’s pioneering hand calculations captured the essence of the problem, while today’s advanced software offers a refined understanding that bridges the gap between theory and field performance.
Figure 3 – Comparison of slip surfaces: Ireland’s circular analysis (FS ≈ 1.1), DeepEX non-circular solution (FS = 0.876), and the observed failure geometry.
Discussion
The Congress Street case provides a fascinating lens through which to examine how analytical approaches in geotechnical engineering have evolved over the past seven decades. Ireland’s original study, carried out with hand calculations in 1954, represents the best of classical soil mechanics: a φ = 0 circular slip surface analysis based on carefully corrected undrained shear strengths. His work demonstrated that the slope was on the verge of instability, with a computed factor of safety of about 1.1.
However, Ireland’s analysis also highlighted a recurring limitation of circular methods. While the calculated factor of safety was in excellent agreement with the imminent failure, the predicted slip surface did not replicate the geometry of the actual slide. His critical circle extended farther back into the slope than what was observed in the field. In essence, his analysis captured the “how safe” but only approximated the “how it fails.”
When the same case is revisited using DeepEX with the Morgenstern-Price method for non-circular surfaces, the picture becomes sharper. The computed factor of safety drops to 0.876, clearly signalling instability rather than marginal stability, a result that better reflects the fact that a failure did occur during excavation. Just as important as the numerical outcome is the geometry: the slip surface predicted by DeepEX initiates a few feet behind the crest, rather than far back into the slope, and daylighted close to the toe of the excavation. This trajectory mirrors the actual field observations much more closely.
The side-by-side comparison of these three perspectives (Ireland’s hand-calculated circle, the non-circular DeepEX solution, and the observed slide) illustrates both the strengths and the boundaries of classical methods. Ireland’s analysis was remarkably accurate given the tools of his time, and his result still stands as a benchmark. DeepEX, by relaxing the constraint of circular slip surfaces, refines the prediction, providing both a more realistic factor of safety and a slip surface geometry that matches the failure mechanism.
This case also reinforces a broader lesson in geotechnical practice: progressive failure and fissured clays demand careful consideration. As Skempton and Hutchinson later emphasized, apparent laboratory strengths often overestimate the mobilized strength in the field. The Congress Street slide is a classic example of this phenomenon, and the consistency between Ireland’s corrected values, the observed failure, and the modern DeepEX analysis highlights the enduring relevance of this insight.
Conclusion
Revisiting the Congress Street open cut demonstrates how historical case records and modern analysis tools can complement one another. Ireland’s 1954 hand analysis, limited to circular slip surfaces, yielded a factor of safety of about 1.1, remarkably close to the threshold of failure and a testament to the robustness of classical φ = 0 methods.
DeepEX, applying the Morgenstern-Price method with non-circular slip surfaces, produced a factor of safety of 0.876, in line with the reality that the slope did fail. More significantly, the geometry of the predicted slip surface was in closer agreement with the observed failure mechanism, starting only a few feet behind the crest and daylighting near the toe.
This dual perspective highlights two essential takeaways:
· Ireland’s pioneering work remains valid and insightful, capturing the essence of the problem with remarkable accuracy for its time.
· Modern computational tools refine our understanding, producing results that not only confirm Ireland’s stability margin but also replicate the geometry of failure observed in the field.
The Congress Street case therefore continues to serve as a benchmark in geotechnical engineering. It shows that careful field documentation and rigorous analysis from the past still guide today’s practice, while advanced tools like DeepEX allow us to revisit these historic cases with new clarity. The dialogue between past and present is a reminder that progress in geotechnical engineering is not a replacement of old knowledge, but a deepening and refinement of it.
References
Ireland, H. O. (1954). Stability Analysis of the Congress Street Open Cut in Chicago. Géotechnique, 4(4), 163–168.
Morgenstern, N. R., & Price, V. E. (1965). The analysis of the stability of general slip surfaces. Géotechnique, 15(1), 79–93.
Skempton, A. W., & Hutchinson, J. N. (1969). Stability of natural slopes and embankment foundations. Proceedings of the 7th International Conference on Soil Mechanics and Foundation Engineering, Mexico City, State-of-the-Art Volume, 291–340.
Let us show you how to reduce your design time by up to 90%!
