Development of Two Empirical Correlations for Tunnel Squeezing Prediction Using Binary Logistic Regression and Linear Discriminant Analysis

Abstract

Squeezing as a large time-dependent deformation can result in irreparable damages for tunneling projects. The accurate prediction of this phenomenon in preliminary stages of tunneling projects has a remarkable role on reducing its destructive effects. In this paper, two new empirical correlations have been presented for squeezing prediction before starting the tunneling project using binary logistic regression (BLR) and linear discriminant analysis (LDA). These correlations have been developed based on a comprehensive database including 220 tunneling case histories. In both correlations, overburden depth (H) and rock mass quality (Q) are the independent variables and squeezing conditions can be predicted as the dependent variable. Quality assessment of these correlations indicated that both equations have high performances for squeezing prediction. In comparison to previously developed empirical equations, proposed equations have led to improvement of prediction capacity. The validation results reveal that LDA and BLR equations are better than the previously developed equations.

Appendix 1: Database for Development of BLR and LDA Correlations (Shafiei et al. 2012; Feng and Jimenez 2015)

Case No.	Depth of cover (H, m)	Rock mass quality (Q)	Squeezing condition
1	225	3.6	0
2	550	4.5	0
3	300	0.4	0
4	150	0.4	0
5	200	0.4	0
6	250	8.5	0
7	200	0.57	0
8	175	0.84	0
9	250	2.71	0
10	150	1.1	0
11	300	6	0
12	220	0.8	0
13	52	15	0
14	34	15	0
15	225	3.6	0
16	340	1.8	0
17	550	5.1	0
18	98	0.080	0
19	111	0.008	0
20	112	0.060	0
21	212	0.040	0
22	261	0.095	0
23	95	0.065	0
24	126	0.300	0
25	138	0.013	0
26	198	0.140	0
27	130	0.200	0
28	276	0.250	0
29	276	0.280	0
30	140	0.009	0
31	300	0.090	0
32	300	0.050	0
33	158	0.230	0
34	112	0.008	0
35	225	0.140	0
36	218	0.070	0
37	114	0.470	0
38	114	0.600	0
39	112	0.008	0
40	250	2.700	0
41	300	1.900	0
42	400	0.512	0
43	181	1.250	0
44	110	0.046	0
45	110	1.000	0
46	140	0.215	0
47	140	2.154	0
48	80	93.500	0
49	190	7.450	0
50	130	1.530	0
51	80	10.000	0
52	500	21.544	0
53	30	0.197	0
54	60	0.021	0
55	30	0.004	0
56	180	4.100	0
57	180	2.200	0
58	160	1.500	0
59	200	5.000	0
60	200	1.000	0
61	160	2.000	0
62	200	0.513	0
63	400	4.140	0
64	490	13.100	0
65	570	3.030	0
66	1217	0.263	0
67	1270	0.367	0
68	1226	0.459	0
69	101	0.067	0
70	115	0.167	0
71	102	0.592	0
72	101	1.666	0
73	101	16.657	0
74	101	51.593	0
75	153	37.272	0
76	261	30.956	0
77	261	4.947	0
78	365	5.325	0
79	394	1.562	0
80	157	1.034	0
81	80	1.300	0
82	80	1.100	0
83	50	4.642	0
84	308	0.541	0
85	280	0.05	1
86	280	0.022	1
87	380	0.51	1
88	240	0.12	1
89	300	0.023	1
90	350	0.5	1
91	480	0.8	1
92	410	0.18	1
93	680	0.05	1
94	100	0.010	1
95	100	0.005	1
96	284	0.090	1
97	112	0.006	1
98	800	2.500	1
99	500	0.030	1
100	400	0.030	1
101	285	0.100	1
102	410	0.300	1
103	415	0.880	1
104	500	1.000	1
105	510	0.880	1
106	440	0.050	1
107	450	0.060	1
108	400	0.030	1
109	400	0.050	1
110	200	0.020	1
111	325	0.030	1
112	700	0.300	1
113	550	1.700	1
114	635	4.000	1
115	650	4.120	1
116	450	0.310	1
117	750	0.500	1
118	450	0.590	1
119	337	0.007	1
120	337	0.011	1
121	337	0.006	1
122	337	0.080	1
123	550	0.029	1
124	600	0.023	1
125	600	0.030	1
126	600	0.018	1
127	620	0.020	1
128	620	0.008	1
129	620	0.009	1
130	620	0.016	1
131	620	0.020	1
132	620	0.025	1
133	580	0.023	1
134	580	0.025	1
135	550	0.025	1
136	575	0.007	1
137	700	0.417	1
138	700	0.333	1
139	750	0.333	1
140	600	0.250	1
141	850	0.056	1
142	600	0.033	1
143	300	0.001	1
144	400	0.003	1
145	800	0.194	1
146	300	0.033	1
147	312	0.094	1
148	280	0.083	1
149	270	0.125	1
150	285	0.063	1
151	280	0.031	1
152	280	0.042	1
153	727	2.287	1
154	736	2.426	1
155	733	2.903	1
156	690	1.650	1
157	577	1.517	1
158	200	0.020	1
159	218	0.013	1
160	252	0.010	1
161	246	0.010	1
162	284	0.008	1
163	285	0.008	1
164	211	0.010	1
165	238	0.010	1
166	230	0.015	1
167	223	0.015	1
168	120	0.010	1
169	500	0.215	1
170	500	1.000	1
171	400	0.001	1
172	400	0.046	1
173	650	3.400	1
174	650	0.640	1
175	100	0.100	1
176	650	4.240	1
177	650	0.087	1
178	400	0.211	1
179	100	0.050	1
180	890	0.120	1
181	580	0.236	1
182	250	0.060	1
183	230	0.210	1
184	285	0.004	1
185	102	0.016	1
186	396	0.394	1
187	200	0.173	1
188	183	0.148	1
189	150	0.003	1
190	150	0.316	1
191	31	0.001	1
192	200	0.031	1
193	33	0.001	1
194	46	0.001	1
195	45	0.003	1
196	77	0.001	1
197	53	0.001	1
198	80	0.001	1
199	96	0.002	1
200	100	0.001	1
201	99	0.004	1
202	96	0.008	1
203	74	0.009	1
204	91	0.021	1
205	96	0.018	1
206	139	0.025	1
207	142	0.017	1
208	202	0.004	1
209	199	0.022	1
210	277	0.021	1
211	279	0.005	1
212	399	0.006	1
213	348	0.028	1
214	100	0.004	1
215	300	0.06	1
216	355	0.341	1
217	255	0.04	1
218	180	0.063	1
219	284	0.095	1
220	112	0.006	1

1. 0 non-squeezing and 1 squeezing condition