Purpose: Our project aimed at building an in silico model based on our recently developed in vitro osteoarthritis (OA) model seeking for refinement of the model to enhance validity and translatability towards the more sophisticated simulation of OA. In detail, the previously 3D in vitro model is based on 3D chondrogenic constructs generated solely from human bone marrow derived mesenchymal stromal cells (hMSCs). Besides studying the normal state of the model over 3 weeks, the in vitro model was treated with interleukin-1β (IL-1β) and tumor necrosis factor alpha (TNFα) to mimic an OA-like environment. In order to provide suitable data for the in silico model, image analysis pipelines had to be optimized for determining cell and matrix concentrations. As a subsequent mathematical strategy, we describe the biological processes by differential equations considering, e.g., the change in cell numbers and collagen II concentrations in different areas of the constructs to include spatial resolution over time. Model parameters such as apoptosis, matrix production or degradation rates were calibrated or taken from the literature. Methods: Undifferentiated hMSCs derived from femoral bone marrow samples of patients undergoing total hip replacement were used to generate the chondrogenic constructs via mechanical stimulation. In a first step, the constructs were cultured under normal conditions in cytokine-free medium over a period of 3 weeks. Samples were taken weekly for histology (HE and Alcian blue stainings), immunohistochemistry (Col I and II), protein mass spectrometry and real-time quantitative PCR. To mimic the in vivo inflammatory environment of OA, constructs were treated with IL-1β (50 ng/ml) and TNFα (100 ng/ml) for 3 weeks with medium exchange every 3 days and samples were processed as described above. Images were analyzed using ImageJ with an optimized analysis protocol to extract cell numbers and matrix compositions. The normal state of the constructs (cultivation over 3 weeks untreated) was mathematically modeled with ordinary differential equations (ODE) and the assumptions of a homogeneous spatial distribution based on the in vitro results as well as a basal synthesis and degradation rate of matrix components. Our next step was to use partial differential equations (PDE) to describe the effects after 3 weeks of stimulation assuming an inhomogenous spatial and temporal distribution. Mathematical simulation was performed with KARDOS and Matlab. Results: Our chondrogenic in vitro model revealed the expression of the cartilage specific markers Col II and aggrecan and the deposition of glycosaminoglycans (GAGs) in in the extracellular matrix. The cell concentration was slightly decreased over 3 weeks of untreated cultivation. After stimulation with inflammatory cytokines, the constructs showed an increased expression of inflammatory markers (TNFα, IL-1, -6 and -8) and matrix degrading enzymes (matrix metalloproteinase (MMP)-1, -3 and -10) compared to untreated controls. During histological and histomorphometric analysis, we observed a decreased compactness of extracellular matrix, a loss of Col II, a significantly reduced cell concentration and altered cell morphology. Based on the developed equations (ODE and PDE), we were able to simulate the distribution processes within the constructs including the diffusion from the outside to the inside of especially IL-1β and the corresponding Col II degradation and cell number reduction. Therefore, optimal parameter values were determined by fitting the in silico model data to the in vitro observations. Conclusions: Using the previously developed chondrogenic constructs and stimulating them with pro-inflammatory cytokines allows us to generate a feasible in vitro model for OA. The integration of collected results into the mathematical in silico model is considered to facilitate the refinement of our in vitro model in the future and to plan and determine experiments and outcome parameters more precisely. By combining methods used in biological research and those used in mathematical systems biology, we aim at developing a valid, efficient and attractive alternative approach to test possible treatments for OA, examine underlying mechanism of OA and cartilage repair to further support translation in OA research.