Research on risk assessment method of wireline logging sticking and analysis of sensitive factors Authors: Kai Wei, Boyu Zhao, Wenqiang Feng, Yan Yan, Guoshan Zhao Published: 2026 ABSTRACT AbstractIn this paper, a quantitative evaluation method of wireline logging adhesion during jacking operation is proposed based on the soft string model Firstly, the principle of cable sticking is elucidated in order to judge the cable logging sticking. Then, combined with the soft rod model, the calculation theory of wireline logging adhesion is proposed, and the evaluation model of wireline logging adhesion is established. The model quantifies the risk of wireline sticking, and proposes a risk probability assessment method for wireline sticking during jacking. Finally, a case study of an actual well was conducted to verify the effectiveness of the proposed assessment method. The results show that this well shows signs of wireline logging sticking, and the assessment method is consistent with the actual field observations. An analysis of the sensitive factors influencing cable adhesion and sticking reveals that the cable adhesion and sticking risk is most sensitive to wellbore inclination and drilling fluid density. Considering the combined effects of the main sensitive factors, the overall cable adhesion and sticking risk is reduced by approximately 35%. The results demonstrate that the cable adhesion and sticking evaluation method established in this study is effective, capable of assessing the risk of cable adhesion and sticking in new wells. The method is scientifically sound and has practical application value.Keywords: Wireline logging, Wireline sticking, Soft string model, Risk probability, Analysis of sensitive factorsSubject terms: Energy science and technology, Engineering, Solid Earth sciences Introduction Wireline logging is a crucial technology for acquiring various petroleum geological and engineering technical data1–4;. In recent years, with the increasing depth of drilling5,6; and the growing number of directional wells7, the risk of wireline logging has gradually increased, and wireline sticking has become a common issue. Prolonged inactivity of logging instruments, poor mud performance, or improper implementation of technical measures may increase the possibility of wireline sticking. Generally, after wireline sticking appears, it can only be determined whether it can be released by lifting and lowering the wireline8. However, the tension that the logging wireline can withstand is limited, and excessive lifting may cause the wireline to pull off, resulting in the instrument sinking to the bottom of the well, Conversely, lowering the wireline may exacerbate ad-hesion issues. Currently, scholars have extensively researched the principle of wireline adsorption sticking. Ruslan Manakhayev9 analyzed Tengiz Oilfield and found that the mud density is large, and strong positive pressure difference will be generated locally, which makes the well wall adsorb logging instruments and wirelines to a certain extent, affecting the lifting and lowering of logging instruments. Tjioe and Martin10 analysed the main factors responsible for wireline sticking in the deepwater Gulf of Mexico, including settling time, bare hole length, wellbore diameter and drilling fluid performance. It can be found that the predecessors have basically determined the influential factors of wireline sticking11, and given some methods to prevent wireline sticking12, but there is no set of physical calculation methods to determine whether wireline sticking occurs. Current research on wireline logging sticking faces issues such as a lack of quantitative methods, insufficient model universality, unsystematic preventive measures, and limited experimental data. Conclusions are often based on specific regions and theoretical analysis, lacking practical validation. In order to clarify the risk of wireline logging sticking and judge the risk probability of wireline logging sticking, a method of quantitative evaluation of wireline logging sticking is proposed by establishing a physical model to realize the quantitative evaluation of wireline logging sticking. The calculation theory of wireline friction is adopted, and the influence of mud cake thickness on wireline sticking is quantitatively characterized to form a quantitative evaluation method of wireline logging risk, which can clarify the risk probability of wireline adsorption and realize the quantitative evaluation of wireline sticking One at A Time (OAT) method is used to determine the ordering of the main control factors of wireline sticking, which has wide application value. Materials and methods Physical model and hypotheses In order to clarify the risk of wireline logging sticking and judge the risk probability of wireline logging sticking, a method of quantitative evaluation of wireline logging sticking is proposed by establishing a physical model to realize the quantitative evaluation of wireline logging sticking. The calculation theory of wireline friction is adopted, and the influence of mud cake thickness on wireline sticking is quantitatively characterized to form a quantitative evaluation method of wireline logging risk, which can clarify the risk probability of wireline adsorption and realize the quantitative evaluation of wireline sticking One at A Time (OAT) method is used to determine the ordering of the main control factors of wireline sticking, which has wide application value. In the process of wireline logging, the wireline is embedded into the mud cake due to its own weight and pressure difference (the difference between wellbore drilling fluid pressure and formation pressure), etc. When the friction force of the wireline in the mud cake is greater than the safe tensile tension of the wireline, the wireline sticking will be generated. The schematic diagram of wireline sticking is shown in Fig. 1. The schematic diagram of wireline sticking. Drilling fluid can balance the pressure between the borehole and formation to ensure the stability of the borehole wall. But mud cake is easy to form on the well wall of the sandstone section with good permeability13, the porosity and permeability of mudstone are smaller than that of sandstone, and the quality of mud cake is poor, generally no thick mud cake is formed, and wireline sticking is easy to appear in the sandstone section14. Therefore, the analysis of onsite accident cases and investigation shows that wireline sticking is closely related to formation properties, drilling fluid properties and mud cake quality. The following hypotheses we put forward for the modeling: The wireline is similar to a flexible rope, with no bending stiffness but can transmit tension; The wireline is similar to a flexible rope, with no bending stiffness but can transmit tension; The wireline has the same shape as the borehole axis and is in continuous contact with drilling fluid; The wireline has the same shape as the borehole axis and is in continuous contact with drilling fluid; Ignoring the influence of the dynamic circulation of the fluid in the wellbore, it is considered that the drilling fluid is relatively stable during logging15. Ignoring the influence of the dynamic circulation of the fluid in the wellbore, it is considered that the drilling fluid is relatively stable during logging15. This assumption is highly applicable in conventional cable logging operations, as there is typically no circulation established within the wellbore during logging, and the drilling fluid remains largely stationary. Therefore, the additional impact on the cable’s loading can be neglected. Wireline friction calculation soft string model In this study, a soft-string model was employed to characterize the contact and frictional behavior between the wireline and the borehole wall16. Based on the mechanical equilibrium of the wireline and the geometrical relationship of the well trajectory, this model can effectively describe the force distribution of the wireline within the wellbore. On this basis, a mud-cake correction was further introduced, allowing the influence of borehole wall conditions on the friction coefficient to be represented more realistically, thereby directly serving the prediction of adhesion-related sticking risk. The model has a clear physical basis, which facilitates parameter sensitivity analysis to identify the key factors governing sticking. Meanwhile, its computational framework is relatively simple, offering good practicality and computational efficiency. o To more accurately simulate the mechanical behavior of the cable under actual working conditions, the following basic assumptions were introduced during the modeling process to simplify the problem and highlight key physical mechanisms: Cable Mechanical Characteristics Assumption: The cable is assumed to exhibit flexible and non-rigid characteristics under loading, with its deformation being dynamic and directly related to the applied external forces. Cable Mechanical Characteristics Assumption: The cable is assumed to exhibit flexible and non-rigid characteristics under loading, with its deformation being dynamic and directly related to the applied external forces. Cable Bending and Curvature: The curvature of the cable will change in response to external forces. During loading, the cable will not only stretch axially but will also bend, and this bending is closely related to the applied load. Cable Bending and Curvature: The curvature of the cable will change in response to external forces. During loading, the cable will not only stretch axially but will also bend, and this bending is closely related to the applied load. Frictional Forces: The frictional forces between the cable and the contact surface play a critical role during the cable’s loading process. These frictional forces resist the movement of the cable, affecting its deformation and force distribution. Frictional Forces: The frictional forces between the cable and the contact surface play a critical role during the cable’s loading process. These frictional forces resist the movement of the cable, affecting its deformation and force distribution. The wireline is divided into corresponding computing units according to the inclination data or the equinox data. Based on the Lagrange principle and nonlinear beam theory17, the floating weight per unit length of each microelement can be calculated, and the positive contact pressure, friction force and upper axial force of the wireline microelement can be calculated by piecewise calculation, as long as the axial force of the lower end of the microelement is known. The axial force at the lower end of the bottom wireline element (the joint connected with the instrument) is equal to the floating weight of the instrument, and the friction force and axial force distribution of the entire wireline can be calculated from the bottom up18.The wireline’s underground force is shown in Fig. 2. The schematic diagram of wireline force. According to the measured inclination and azimuth data, the wireline in the open-hole section is divided into a series of micro-elements. For an arbitrary micro-element, a local coordinate system is established at the midpoint of the element, where is the tangential unit vector, is the principal normal unit vector, and is the binormal unit vector. Let α1 and α2 denote the inclination angles at the lower and upper ends of the micro-element, respectively, and let Δφ denote the azimuth increment. The dogleg angle is denoted by γ. The tangential unit vectors at the lower and upper ends of the micro-element can be expressed as: . The azimuth increment is defined as . The dogleg angle γ satisfies For a sufficiently short micro-element, the midpoint unit vectors are written as Since the included angle between t1 and t2 is γ, one has Let W denote the buoyant weight of the wireline micro-element. The buoyant weight vector is decomposed in the local coordinate system as Projecting the buoyant weight onto the three local directions gives The calculation formula is as follows: . Because the tangential directions at the two ends form an angle γ, the projections of the end tensions on the midpoint tangential direction are T1cos(γ/2) and T2cos(γ/2), respectively. By taking force equilibrium along the tangential direction et, one obtains In the normal plane formed by (en,eb), the force component in the principal normal direction is,and the force component in the binormal direction is Wb. Hence, the contact normal force can be written as For a sufficiently short micro-element, the tension difference between the two ends is small, and therefore T1 ≈ T2≈T(x). Then Let the upper-end tension be denoted by T(x) and the lower-end tension by T(x-Δx). The tangential equilibrium equation becomes Rearranging and dividing both sides by Δx, and then taking the limΔx→0, Substituting the expression of FN into the tension differential equation yields Assume that the lower-end tension at the bottom of the open-hole section is T(0), and let the total length of the open-hole wireline section be L. Integrating the above equation from x = 0 to x = L gives: Equation (1.2) is the integral expression of the total wireline tension in the open-hole section. Once the lower-end boundary condition T(0) and the wellbore trajectory data are given, the axial-tension distribution and the total wellhead tension can be obtained iteratively from bottom to top. The Calculation Model of Mud Cake Thickness. The Calculation Model of Mud Cake Thickness. The thickness of mud cake is affected by both solid phase deposition and liquid phase erosion of wellbore working fluid19. During the filtration process of wellbore working fluid, the solid particles in the working fluid will be deposited on the surface of wellbore wall to form mud cake. The flow of the working fluid in the annulus will scour the mud cake, causing some particles on the surface of the mud cake to reenter the working fluid and flow with the working fluid20. When the mass balance between solid deposition and liquid phase erosion is reached, it is the actual thickness of mud cake in wireline logging. The accumulation rate of solid particles in the shaft wall can be expressed as: . where Rs is the stacking rate of solid particles on the surface of mud cake, g/(cm2 s); Kd is the sedimentation coefficient of solid particles on the surface of mud cake, S; ρf is the drilling fluid density, g/cm3; µ is the flow rate of the filtrate, cm/s; Bp is the mass fraction of solid particles in the working fluid; ke is the erosion coefficient of the mud cake, g/(N·s); τs is the shear of the working fluid on the mud cake, Pa; τcr is the critical shear stress of the mud cake, Pa. In the process of forming mud cake, it is known from the conservation of solid particle mass that: Where is the porosity of the mud cake, %; ρp is the density of solid phase particles, g/cm3; rb is the radius of the wellbore wall, cm. Substitute Eq. (3) into Eq. (5) and integrate to obtain the mud cake thickness expression: where rw is the initial wellbore radius, cm; rc is the realtime wellbore radius, cm. Since the working fluid will generate shear stress on the mud cake when it returns, if the flow of the working fluid in the annulus is laminar flow, the shear stress can be seen from the RabinowitzMooney equation(Guillot and Colin 2014) as follows: where n is the drilling fluid flow pattern index; K is the viscosity coefficient of drilling fluid, Pa·sn; v is annular reverse velocity of drilling fluid, cm/s; rd is the outer diameter of drill pipe, cm. The porosity, permeability and critical shear stress of mud cake can be measured by experiment21. Assuming that the flow of the working fluid follows Darcy’s law, the flow of the fluid can be divided into three parts, that is, the flow of the working fluid in the mud cake, the flow of the formation and the flow of the crude oil in the formation. The expression of working fluid filtration flow rate is: where q is the seepage flow rate of the mud cake intrusion zone formation, cm3/s; h is the thickness of the formation, cm; Δp is the pressure difference, MPa; µf is the viscosity of drilling fluid, mPa·s; Kf is the cake permeability, µm2; KR is the permeability, µm2; re is the outer boundary of filtrate flow, cm; rD is the radius of filtrate invasion, cm; µ0 is the viscosity of the formation fluid, mPa·s. According to the flow law of the plane radial flow, the flow rate of filtrate at the mud cake can be calculated as follows: The cumulative filtration loss is: where Q is the accumulated filtration loss per unit formation thickness, L; t is the time, min. Equation (8) is discretized by difference, and the difference equation of hole radius when mud cake is formed can be obtained after sorting. The filtration rate of working fluid can be obtained by Eq. (10) and Eq. (11). By iteratively solving the difference equation of Eq. (8) and Eq. (10) and Eq. (11), the mud cake thickness at a given time can be obtained. By iterating the difference equation of Eq. (12), the cumulative filtration loss of the working fluid can be obtained. Modify the Calculation Model of Friction Coefficient. Modify the Calculation Model of Friction Coefficient. Different mud cake deposits will affect the friction coefficient22,23. After determining the thickness of mud cake, the friction coefficient can be modified to a certain extent, and the correction formula for the friction coefficient can be established: where µcw is the corrected friction coefficient; µw is the friction coefficient when the mud cake is not considered; Δµ is the expected maximum increase in the coefficient of friction after taking into account the mud cake; ds is the characteristic mud cake thickness; λ is the exponential rise of the friction coefficient curve. Based on field test data from cable logging, it is known that the friction coefficient for typical cased-hole sections is approximately 0.2, while for open-hole sections, it is approximately 0.35. Since this study primarily focuses on the cable adhesion in open-hole sections, the initial friction coefficient is set to 0.35.On this basis, the effect of mud cake thickness on the friction coefficient is considered, and the initial value is adjusted: when mud cake is not considered, the friction coefficient is 0.35; when mud cake is present, the friction coefficient is adjusted according to the change in mud cake thickness. The difference equation from Eq. (10) is used, combined with Eqs. (12) and (13), to iteratively solve for a characteristic mud cake thickness of 0.09 cm. Accordingly, as depth increases, the friction coefficient follows a variation pattern controlled by the mud cake thickness, with the curve’s rising exponent being − 1. The corrected friction coefficient is shown in Fig. 3. The blue is the friction coefficient before correction, and the crocus is the friction coefficient after correction. The graph of corrected friction coefficient. The Calculation Model of Adsorption Friction. The Calculation Model of Adsorption Friction. According to Hertz theory24, when the wireline is in contact with the mud cake, analogous to the Hertz model of cylindrical contact in Hertz contact, as shown in the Fig. 4. The diagram of Hertz contact. The positive pressure is linearly proportional to the penetration depth: Equivalent elastic modulus: Equivalent radius: Contact half length: Contact perimeter: Contact area: Additional contact friction: Therefore, the formula of contact friction between wireline and mud cake based on Hertz basis can be established. The Calculation Model of Risk Probability of Wireline Sticking. The Calculation Model of Risk Probability of Wireline Sticking. The occurrence of wireline differential sticking is essentially governed by whether the total surface tension exceeds the critical tension, Tc. During actual field operations, the total surface tension is not constant; rather, it varies between the minimum total tension, Tmin, and the maximum total tension, Tmax, owing to the combined effects of wellbore trajectory, frictional drag, downhole fluid conditions, and wireline operating parameters. Therefore, the interval [Tmin,Tmax]can be regarded as the possible fluctuation range of the total surface tension, while the interval [Tc, Tmax], in which the critical tension is exceeded, may be defined as the hazardous range for wireline differential sticking. Based on this consideration, the risk of wireline differential sticking is characterized by the ratio of the hazardous interval to the overall variation range of total surface tension. Accordingly, the probability of wireline differential sticking is defined as: The closer the critical tension Tc is to the minimum total tension Tmin within the total tension variation interval, the larger the hazardous interval in which the total tension exceeds the critical threshold, and consequently, the higher the likelihood of wireline differential sticking. Conversely, as Tc approaches the maximum total tension Tmax, the hazardous interval becomes smaller, and the risk of differential sticking correspondingly decreases. Finally, the whole calculation process is solved and the results are displayed. The calculation flow chart of wireline sticking is shown in Fig. 5. Wireline sticking calculation flow chart. Results and discussion Model validation Among the numerous simulation data, a case well drilled was selected, and the cable adhesion evaluation model was used to analyze the adhesion and sticking risk during its electric logging operation. The simulation parameter values for this well are shown in Table 1. It is a directional well with a drilling fluid density of 1.35 g/cm³, a drilling fluid viscosity of 54 s, a drill bit diameter of 219.9 mm, a technical casing inner diameter of 244.47 mm, and a maximum wellbore inclination of 26.72°. The length of the logging instrument used is 15.9 m, with a weight of 735.6 kg and an outer diameter of 152.4 mm. During the rotary wall coring operation, cable adhesion and sticking occurred. A total of 25 cores were planned to be taken, but sticking was encountered after coring at around 4100 m. The on-site tension curve is shown in Fig. 6, where TTEN is the total tension, CHT is the bumper tension, and TEN is the differential tension. Both TTEN and TEN increased, while CHT remained constant. The data analysis results indicate that the issue was cable adhesion and sticking. Table 1The values of the simulated parameter variables.ParameterValueParameterValueDrilling fluid density, (ρf) g/cm31.35Drilling fluid viscosity (µf), mPa·s54Bit diameter, mm215.9Technical casing hole diameter, mm244.47Maximum hole slope, °26.72Instrument length, m15.9Instrument quality, kg735.6Instrument outside diameter, mm152.4Mud cake porosity (), %0.1Solid particle deposition coefficient (Kd)0.01Mud cake erosion coefficient(ke)1Critical shear stress of mud cake (τcr), Pa22.5Mud cake permeability (Kf), dc0.05Drilling fluid viscosity coefficient (K)0.8Drilling fluid flow pattern index0.6Mass fraction of solid particles (Bp)0.1Solid particle density (ρp), g/cm31.8Bare eye Friction coefficient (f)0.35Technical casing down depth, m3400Casing friction coefficient (f)0.2 The values of the simulated parameter variables. The diagram of field tension. The wireline adsorption evaluation model was used to calculate the well. The mud cake thickness was shown in Fig. 7(a), the corrected friction coefficient was shown in Fig. 7(b), and the risk probability of wireline sticking was shown in Fig. 8. The results from the figure show that the model’s predicted probability of cable adhesion and sticking risk for this well is generally higher than 80%, indicating a very high risk level. In the actual field operation, cable adhesion and sticking did occur in this well, which is highly consistent with the model’s prediction. This comparison demonstrates that the model not only reasonably reflects the cable’s force and frictional state under wellbore conditions, but also maintains strong applicability and stability under complex geological conditions. The model has high accuracy and engineering guidance value for the quantitative assessment of cable adhesion risks, providing reliable technical support for on-site operations. (a) The graph of mud cake thickness, (b) The graph of corrected friction coefficient. The risk probability diagram of wireline sticking. Parameter sensitivity analysis One of the simplest and most common methods in sensitivity analysis is the OAT method, which typically involves moving one input variable, keeping the baseline values of the other variables, then returning the variable to its nominal value, and then repeating the same way for each other input. After that, the sensitivity can be measured by monitoring changes in the output. The OAT method can clearly view the influence of a single sensitivity factor on the output result, which is intuitive and reliable. Therefore, in order to clearly display the sensitivity factors of wireline sticking, OAT analysis is used to change only one parameter at a time to see its impact on the risk probability, so as to evaluate the risk factors of wireline sticking scientifically and reasonably. The local analysis method is to perform parameter sensitivity analysis25 on a point in a multidimensional parameter space, which is assigned a default value or manually rough ratio. Sensitivity is usually obtained by calculating the partial derivative of the output function related to the parameter. The calculation method for parameter sensitivity S is shown in formula (20). S can be obtained by changing the value of Δe while keeping other parameters unchanged: where M is the model output; ei represents different model parameters; Δei represents a small disturbance to the ei parameter. The first order second order moment method estimates the average (first order second order origin moment) and standard deviation (second order center moment) of the model output by calculating the derivative of a certain point. The steps of the first order second order moment method are to first select the probability distribution of the theoretical distribution family g(y) = g(y,θ) approaching M = f(e1,e2,···,en), and then use the Taylor formula to expand M at the mean of (e1,e2,···,en), discarding higher order terms of more than two degrees. This approximates the second order moment of M and estimates the parameters. The first-order term of Taylor’s formula needs to calculate the model output at a certain point and determine its derivative: where is a random input variable related to the mean ; δM is the derivative value of δe at the mean .The mean and standard deviation σM of output function M can be calculated using the following equation: The relevant parameters were sorted out, and OAT analysis was applied to well inclination, porosity, drilling fluid density and viscosity, and these parameters were changed successively to observe and analyze the changes in the probability of wireline sticking. (1) Well inclination angle. The inclination angle, defined as the angle between the wellbore axis and the direction of gravity, directly affects the force state of the wireline within the wellbore. Based on the force analysis of the soft-string model, as the inclination angle increases, the component of the wireline weight acting normal to the borehole wall correspondingly increases, resulting in a significant rise in the normal contact force between the wireline and the borehole wall. Meanwhile, the contact length of the wireline along the borehole wall also increases, further expanding the frictional interaction zone. According to the classical friction law, in which friction force is proportional to the normal force, these changes lead to an increase in the overall drag acting on the wireline during its movement in the wellbore, thereby aggravating the risk of adhesion-related wireline sticking. Therefore, the inclination angle is identified as a key parameter influencing the risk probability, which increases monotonically with increasing inclination angle. (2) Porosity. Porosity refers to the ratio of the volume of interconnected pores in a rock to the total volume of the rock. Based on the single factor analysis of porosity, it can be found that porosity has little effect on the risk of adsorption card. (3) Drilling fluid density. The selection of drilling fluid density should comprehensively account for both formation pressure balance and wellbore stability requirements. An increase in drilling fluid density directly raises the hydrostatic pressure within the wellbore. On the one hand, this enhances the compaction of the mud cake on the borehole wall, thereby increasing the friction coefficient at the mud-cake surface. On the other hand, if the formation pore pressure remains unchanged, the pressure differential between the wellbore and the formation correspondingly increases, resulting in a greater lateral extrusion force acting on the wireline. These two effects jointly increase the contact pressure and frictional resistance between the wireline and the borehole wall, thereby significantly elevating the probability of adhesion-related sticking. Therefore, drilling fluid density is identified as another key influencing factor, and the corresponding risk probability increases with increasing drilling fluid density. (4) Drilling fluid viscosity. Drilling fluid viscosity represents the overall manifestation of internal friction arising from interactions between suspended solid particles and liquid molecules during fluid flow. A high-viscosity drilling fluid can form a stable fluid film within the annular gap between the wireline and the borehole wall, thereby achieving solid–solid separation through the hydrodynamic lubrication effect and significantly reducing sliding friction. At the same time, during the squeezing process, a higher-viscosity fluid generates greater resistance, which delays the approach of the wireline toward the borehole wall and mitigates the impact of normal contact. In addition, high-viscosity drilling fluid is beneficial for filtrate-loss control and promotes the formation of a dense and lubricating thin mud cake, thereby preventing the development of localized high-friction points. Therefore, the probability of adhesion-related wireline sticking decreases with increasing drilling fluid viscosity. Figure 9(a) is the risk probability diagram after the change of well inclination, Fig. 9(b) is the risk probability diagram after the change of porosity, Fig. 10(a) is the risk probability diagram after the change of drilling fluid density, and Fig. 10(b) is the risk probability diagram after the change of viscosity. After the well inclination Angle, porosity, drilling fluid density and viscosity are reduced by 10%, it is found that the risk probability has different changes. Among them, the risk probability decreases by about 30% after the well inclination angle is reduced by 10%, and the risk probability decreases by about 40% after the drilling fluid density is reduced by 10%. These two factors have a great impact on the risk probability of cable adsorption and are the key factors. Porosity and viscosity have some effects, but the effect is relatively insignificant. Therefore, through OAT analysis, it can be seen that the main factors affecting wireline logging sticking are the size of the well inclination Angle and the size of the drilling fluid density, both of which are positively correlated, that is, within a reasonable range, reducing the size of the well inclination Angle or reducing the drilling fluid density will reduce the risk probability of wireline logging sticking during wireline logging. Therefore, when designing the well structure, we should fully consider the two situations and reasonably design the new well. (a) The risk probability diagram after the change of well inclination, (b) The risk probability diagram after the change of porosity. (a) The risk probability diagram after the change of drilling fluid density, (b) The risk probability diagram after the change of viscosity. Applicability analysis Compared to existing research methods, the physical model developed in this study has clear advantages in terms of mechanistic explanation and parameter controllability. Unlike other methods that primarily rely on large amounts of historical well data for training and fitting, the model in this study is built directly on the mechanical mechanisms of cable-wellbore interaction, providing a clear explanation of the physical causes behind adhesion and sticking risks. Additionally, the input parameters of this model are measurable or controllable engineering parameters (such as wellbore inclination, porosity, drilling fluid density, and viscosity), allowing researchers and engineers to quantitatively assess the contribution of different factors to the risk through sensitivity analysis. It should be noted that, in new wells or complex well sections with limited historical data, the physical model has a greater advantage, while in well areas with sufficient data accumulation, data-driven methods may offer superior predictive efficiency. Therefore, in terms of technical positioning, the model developed in this study is well-suited as a mechanism-driven prediction tool, providing reference for early-stage risk assessment and parameter optimization in on-site operations. To verify the effectiveness and generalization capability of the proposed model, a systematic validation was conducted in this study using field logging data from a total of 38 wells in the South China Sea. One directional well in the South China Sea was selected as a representative case for detailed analysis, with the aim of clearly illustrating the model development procedure and application workflow. The case selection was intended to ensure both typicality and representativeness, rather than to confine the applicability of the model to a single-well condition. From the perspective of the underlying physical mechanism, the model was developed on the basis of wireline mechanical equilibrium and frictional contact theory. Its governing equations do not depend on a specific well type or geographical region, and therefore the model possesses good theoretical generality. Under different operating environments, such as high-temperature/high-pressure (HTHP) wells, deepwater wells, and wells in cold regions, the fundamental force mechanism and frictional behavior between the wireline and the borehole wall remain essentially unchanged, and the model framework is still applicable. It should be noted, however, that in actual operating conditions, key parameters such as the friction coefficient, drilling fluid rheological properties, and the mechanical properties of the wireline material may vary significantly with environmental factors including temperature and pressure. Therefore, in practical applications, these model parameters should be appropriately calibrated according to specific operating conditions so as to accurately characterize the risk of adhesion-related sticking under different downhole environments. Conclusions Wireline logging is a key technology to accurately obtain formation information, and there is a construction risk of wireline logging sticking in the logging process. According to the phenomenon of wireline logging sticking in the actual logging site, the risk factors of wireline logging sticking were analyzed, and a method for quantitative evaluation of wireline logging sticking risk by establishing a physical model was proposed, which took into account the influence of wellbore mud cake. By modifying the friction coefficient, the calculation method of wireline logging sticking friction is obtained, the risk probability of wireline sticking can be defined, and the quantitative evaluation of wireline logging sticking can be realized. OAT method was used to analyze the sensitivity factors of wireline logging sticking. It was found that inclination angle and drilling fluid density were the main sensitivity factors of wireline logging sticking, and both of them were positively correlated with wireline logging sticking. Reducing inclination Angle could reduce the risk by about 30%, and reducing drilling fluid density could reduce the risk probability by about 40%. In the design of well structure, the two values should be considered reasonably to reduce the risk of wireline logging sticking. In applications across different oilfield regions, the predictive accuracy of the model is affected by factors such as formation architecture, wireline type, mud system, and operating parameters. Under conditions of complex geology, limited historical data, or operating conditions that fall outside the distribution range of the modeling samples, the model may exhibit significant prediction errors. Therefore, the model should be calibrated for specific operating regions through targeted experiments, with key parameters identified and tuned and its applicability limits clearly defined, so as to improve its prediction reliability and generalization capability across different regions and operating conditions. The wireline differential sticking risk prediction model proposed in this study features low computational complexity and strong portability. In the next stage, a regional parameter self-adaptation mechanism will be developed based on this model, enabling rapid calibration of key parameters using a limited amount of experimental data or historical inversion results, thereby reducing the cost of repeated laboratory testing. In parallel, the model will be integrated with lightweight data-driven methods to establish a hybrid inference framework suitable for downhole edge-computing units, allowing online model updating and real-time risk warning, and further enhancing the intelligence level and risk response capability of wireline logging operations. List of symbols the floating weight of the wireline measuring point,N tangential floating weight component, N the normal floating weight component, N the radial floating weight component, N the angle of inclination at the end point of the wireline unit,° the angle of inclination at the lower end of the wireline unit,° azimuth increment, ° dogleg severity, ° the wireline element contact friction force,N the wireline element adsorption friction force,N the positive contact pressure of the wireline element,N the friction coefficient the axial force on the upper end of the wireline element,N the axial force at the lower end of the wireline element,N the stacking rate of solid particles on the surface of mud cake,g/(cm2 s) the sedimentation coefficient of solid particles on the surface of mud cake,S drilling fluid density,g/cm3 the flow rate of the filtrate,cm/s the mass fraction of solid particles in the working fluid the erosion coefficient of the mud cake,g/(N·s) the shear of the working fluid on the mud cake,Pa the critical shear stress of the mud cake,Pa the porosity of the mud cake,% the density of solid phase particles,g/cm3 the radius of the wellbore wall,cm the initial wellbore radius,cm the realtime wellbore radius,cm the drilling fluid flow pattern index the viscosity coefficient of drilling fluid, Pa·sn annular reverse velocity of drilling fluid, cm/s the outer diameter of drill pipe, cm the seepage flow rate of the mud cake intrusion zone formation,cm3/s the thickness of the formation, cm the pressure difference, MPa the viscosity of drilling fluid, mPa·s the cake permeability,μm2 the permeability,μm2 the outer boundary of filtrate flow,cm the radius of filtrate invasion,cm the viscosity of the formation fluid,mPa·s the accumulated filtration loss per unit formation thickness,L the corrected friction coefficient the friction coefficient when the mud cake is not considered the expected maximum increase in the coefficient of friction after taking into account the mud cake the characteristic mud cake thickness,cm the exponential rise of the friction coefficient curve the elastic modulus of the wireline, pa the elastic modulus of the mud cake, pa the radius of the wireline,cm the radius of the hole,cm the penetration depth,cm the sticking length,m the equivalent elastic modulus, pa the equivalent radius,cm the equivalent half length,cm the contact circumference,cm the contact area,cm2 the maximum value of the total tension, N the minimum value of the total tension, N the tension threshold, based on the safety tension standards provided by the cable manufacturer, N the risk probability of the wireline adsorption card, % the model output represents different model parameters represents a small disturbance to the ei parameter random input variable related to the mean the derivative value of δe at the mean Author contributions Kai Wei: Methodology, Writing–original draft, Writing–review and editing.Boyu Zhao: Conceptualization, Programming, Writing–original draft.Wenqiang Feng: Programming, Data collection.Yan Yan: Conceptualization, Data collection.Guoshan Zhao: Conceptualization, Data analysis. Funding This work received funding from the “Scientific research project of Sinopec Jingwei Co., LTD: Development and Application of a Flexible Measurement-While-Drilling (MWD) Tool for Small-Borehole Sidetracking and Short-Radius Wells’ (JWKT2407); Study on a High-Speed Transmission Rotary Valve Pulser Based on Optimized Composite Coding(JWKT2408)”. Data availability The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Kai Wei, Email: weikai630@163.com. Boyu Zhao, Email: zhaoboyu98@163.com. References Associated Data Data Citations Data Availability Statement The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.