Batch crystallization is commonly used in pharmaceutical, agrochemical, specialty and fine chemicals industry. The advantages of batch crystallization lie in its ease of operation and the relatively simple equipment that can be used. On the other hand a major disadvantage associated with it is the inconsistent and usually poor product quality. Quality of the crystalline product, which is defined in terms of the Crystal Size Distribution (CSD), purity, kind of solid state etc., is related to its performance when used as an ingredient during subsequent processes. Also the quality of the product from batch crystallization process has a strong influence on the efficiency of downstream operations like filtration and drying. Hence it is essential to reduce the batch-to-batch variations in the product quality. In this thesis three basic requirements for achieving consistent product quality have been identified. These requirements are a.) strong domain knowledge, b.) proper means of characterizing crystallization phenomena, c.) adequate process monitoring capabilities. The results presented in this thesis help in meeting the above requirements and are summarized below. a. Crystallization domain knowledge: Three important results related to the Metastable Zone Width (MSZW) have been obtained in this thesis which cannot be explained by the conventional understanding. It has been shown in this thesis that i. MSZW is not a deterministic property ii. MSZW is volume dependent iii. There exists a relationship between MSZWs measured at different volumes under similar conditions. The MSZW measurements at small volumes of 1 mL show large variations while the variations in the measurements reduce as the volume is increased. The extent of variations in the MSZW measurements at a particular volume changes from one model system to the other. The smallest measured MSZW at all volumes between 1 mililitre and 1 litre is the same. The dramatic deviation from the conventional understanding of the measured MSZWs is a result of inadequate understanding of the nucleation process. Conventionally, a multiple nuclei mechanism is assumed in which large number of nuclei are born together in a very short time interval. However in this thesis evidence is presented for a mechanism in which only a single nucleus is formed initially in a supersaturated solution which grows into a single crystal. After growth to a certain size, this single crystal undergoes extensive secondary nucleation which results into multiple crystalline fragments. The newly postulated mechanism is called the Single Nucleus Mechanism. All the crystals produced in an unseeded batch crystallization therefore originate from a single primary nucleus by secondary nucleation. This indicates that during an unseeded industrial batch crystallization process, there will be different generations of crystals present. Hence, in order to achieve crystals with desirable quality, control strategies must be focused on controlling both primary and secondary nucleation. b. Crystallization characterization: In this thesis novel methods to characterize crystal nucleation, growth and MSZW have been developed. The characterization of crystal nucleation and MSZW is done with the help of a stochastic model developed based on the Single Nucleus Mechanism. The stochastic model indicates that the nucleation rate is several orders of magnitude smaller than that postulated by the Classical Nucleation Theory. The low nucleation rate leads to the stochastic MSZWs. Unlike the conventional population balance model which shows that the MSZW is independent of volume, the stochastic model indicates that the MSZW is a function of volume. The stochastic model also enables scale dependent study of the MSZW. The characterization of crystal growth is performed by the combination of information from both the concentration measurement sensor and the crystal size distribution (CSD) measurement sensor. It is shown that by combining of the concentration and CSD measurements a better parameter estimation and better process description could be achieved. c. Crystallization process monitoring: In this thesis in situ measurement of several process variables has been successfully demonstrated not only at lab scale but also at industrial scale. A comparison has been performed between two spectroscopy based techniques viz. attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR) and Fourier transform near infrared spectroscopy (FT-NIR) for in situ concentration monitoring during crystallization at lab scale. Based on the comparison, ATR-FTIR is found to be more accurate than FT-NIR for different model systems. In spite of accurate concentration monitoring at lab scale, the concentration monitoring with ATR-FTIR leads to biased measurements at industrial scale due to the differences in the curvature of fiber optics. To facilitate the in situ concentration measurements in industrial environment, two calibration procedures have been investigated which circumvent problems associated with calibration transfer from lab to industrial scale. In the first procedure data from a cheap ultrasound based concentration probe is combined with the spectra from ATR-FTIR spectroscope. It is shown that this combination of data enables a rapid calibration of ATR-FTIR at industrial scale. In the second procedure, multiple Process Analytical Technology (PAT) tools that were arranged in a measurement skid were calibrated simultaneously at industrial scale. The skid configuration of the PAT tools allows for the combination of the calibration procedure with process characterization. The monitoring of the process at industrial scale with multiple sensors brings new process insights which can lead to better process control and optimization strategies. The results presented in this thesis will enable achievement of consistent product quality by facilitating efficient process and equipment design, process development, and process control.