Tool For Process Development Research Articles

A combination of QFD and imprecise DEA with enhanced Russell graph measure: A case study in healthcare

Quality function deployment (QFD) is a proven tool for process and product development, which translates the voice of customer (VoC) into engineering characteristics (EC), and prioritizes the ECs, in terms of customer's requirements. Traditionally, QFD rates the design requirements (DRs) with respect to customer needs, and aggregates the ratings to get relative importance scores of DRs. An increasing number of studies stress on the need to incorporate additional factors, such as cost and environmental impact, while calculating the relative importance of DRs. However, there is a paucity of methodologies for deriving the relative importance of DRs when several additional factors are considered. Ramanathan and Yunfeng [43] proved that the relative importance values computed by data envelopment analysis (DEA) coincide with traditional QFD calculations when only the ratings of DRs with respect to customer needs are considered, and only one additional factor, namely cost, is considered. Also, Kamvysi et al. [27] discussed the combination of QFD with analytic hierarchy process–analytic network process (AHP–ANP) and DEAHP–DEANP methodologies to prioritize selection criteria in a service context. The objective of this paper is to propose a QFD–imprecise enhanced Russell graph measure (QFD–IERGM) for incorporating the criteria such as cost of services and implementation easiness in QFD. Proposed model is applied in an Iranian hospital.

Investigation of the physiological response to oxygen limited process conditions of Pichia pastoris Mut+ strain using a two-compartment scale-down system

Inhomogeneities in production-scale bioreactors influence microbial growth and product quality due to insufficient mixing and mass transfer. For this reason, lots of efforts are being made to investigate the effects of gradients that impose stress in large-scale reactors in laboratory scale. We have implemented a scale-down model which allows separating a homogeneous part, a stirred tank reactor (STR), and a plug flow reactor (PFR) which mimics the inhomogeneous regimes of the large-scale fermenters. This scale-down model shows solutions to trigger oxygen limited conditions in the PFR part of the scale-down setup for physiological analysis. The goal of the study was to investigate the scale-up relevant physiological responses of Pichia pastoris strain to oxygen limited process conditions in the above mentioned two-compartment bioreactor setup. Experimental results with non-induced cultures show that the specific growth rate significantly decreased with increasing the exposure time to oxygen limitation. In parallel more by-products were produced. Examining physiological scalable key parameters, multivariate data analyses solely using on-line data revealed that different exposures to the oxygen limitation significantly affected the culture performance. This work with the small scale-downs setup reflects new approaches for a valuable process development tool for accelerating strain characterization or for verifying CFD simulations of large-scale bioreactors. As a novel methodological achievement, the combination of the two-compartment scale-down system with the proposed multivariate techniques of solely using on-line data is a valuable tool for recognition of stress effects on the culture performance for physiological bioprocess scale-up issues.

Defining Cancer Care Quality or Delivering Quality Cancer Care?

Many in the oncology world see as accepted that quality and value are difficult to define and that no specialty-wide consensus is likely to be reached in the short term about how to measure and report success at improving either. Others find the definition just beyond their grasp and suggest that, like pornography, they know it when they see it. Large organizations, publications, conferences, and processes have sprung up around the concept, and various attempts have been made to develop “measures of quality” that can characterize a practice or physician in this area. This work includes ongoing efforts within and among the National Quality Forum, NCCN, ASCO, The US Oncology Network (The Network), the Community Oncology Alliance, and Ontario Cancer Care to identify quality measures—many of which focus on surrogate processes or activities that should lead to quality. Instead, however, we believe this effort and investment should focus on outcomebased measures, and we offer a paradigm for measuring interventions: •� Quality is the efficient delivery of evidence-based care by trained clinicians in an accessible setting. •� In this context, value is providing higher quality care at the same cost, or the same quality at a lower cost. We believe these concepts are at the core of cancer care delivery, whether in the community or academic setting and across specialties, including medical oncology, radiation oncology, gynecologic oncology, urology, and surgery. Achieving quality will require integrated, coordinated care with clinical teams spanning these professionals and settings. It requires not only coordination among clinicians of different specialties but also investment in infrastructure resources to facilitate the integration of process development and clinical decision support tools that support evidence-based practices that optimize practice efficiency and care delivery. If the right investments have been made and communications highways created, these definitions are measurable, reportable, and comparable (without significant manual effort). Just as important, these definitions allow quality and value to evolve over time as the evidence evolves. Consequently, the highest-quality treatment pathways, and even the best modalities of treatment, may be vastly different in the future. For example, advance care planning and palliative care services would likely not have been included in the concept of evidence-based care 10 years ago. Today, however, the data suggest that palliative care concurrent with disease-directed care can improve outcomes and patient experiences while reducing costs. This view is also supported by a decade of hypothesis, consensus-building, technology investment, data collection, and analysis by physicians and supporting clinicians and staff of The Network, who we believe pioneered the concepts of narrowly drawn, evidence-based pathways for care in oncology. Nearly a decade ago, physicians in The Network decided to develop Level I Pathways, evidence-based guidelines that redirect the wide range of treatments in oncology care into more precise, clinically proven treatment options. (More information is available at http://www.usoncology.com/cancercareadvocates/ AdvancingCancerCare/DeliverHigh-QualityCare/LevelIPathways.) The Level I Pathways program helps physicians eliminate disparities in care by refining potential

Microscale Freeze-Drying with Raman Spectroscopy as a Tool for Process Development

Until recently, the freeze-drying process and formulation development have suffered from a lack of microscale analytical tools. Using such an analytical tool should decrease the required sample volume and also shorten the duration of the experiment compared to a laboratory scale setup. This study evaluated the applicability of Raman spectroscopy for in-line monitoring of a microscale freeze-drying process. The effect of cooling rate and annealing step on the solid-state formation of mannitol was studied. Raman spectra were subjected to principal component analysis to gain a qualitative understanding of the process behavior. In addition, mannitol solid-state form ratios were semiquantitatively analyzed during the process with a classical least-squares regression. A standard cooling rate of 1 °C/min with or without an annealing step at -10 °C resulted in a mixture of α, β, δ, and amorphous forms of mannitol. However, a standard cooling rate induced the formation of mannitol hemihydrate, and a secondary drying temperature of +60 °C was required to transform the hemihydrate form to the more stable anhydrous polymorphs. A fast cooling rate of 10 °C/min mainly produced δ and amorphous forms of mannitol, regardless of annealing. These results are consistent with those from larger scale equipment. In-line monitoring the solid-state form of a sample is feasible with a Raman spectrometer coupled microscale freeze-drying stage. These results demonstrate the utility of a rapid, in-line, low sample volume method for the semiquantitative analysis of the process and formulation development of freeze-dried products on the microscale.

A New Mobility Extraction Technique Based on Simultaneous Ultrafast $I_{d}$–$V_{g}$ and $C_{\rm cg}$–$V_{g}$ Measurements in MOSFETs

Channel carrier mobility in MOSFETs is a key parameter for process development, material selection, and device modeling. The existing techniques for mobility evaluation suffer from one or more of the following shortcomings: slow speed and vulnerability to fast trapping, drain bias dependence, cable-changing, sensitivity to gate leakage, complex procedure, and a need for simulation. This paper proposes and develops a new technique to overcome these shortcomings. - and - are simultaneously measured so that the effect of on mobility is inherently taken into account, and the measured mobility becomes -independent. The cable connection switching between - and - measurements is avoided, and the measurement completes in one pulse. This allows the measurement time reducing to the order of microseconds and, in turn, minimizing the effect of charge trapping. Unlike the standard high-frequency - , is independent of gate leakage here, and the applicability of new technique to thin gate oxides of high gate leakage will be demonstrated. These advantages, together with its easy implementation, should make this technique a simple and robust tool for process development, material selection, and device modeling in future generations of CMOS technology.

A rapid method for the differentiation of yeast cells grown under carbon and nitrogen-limited conditions by means of partial least squares discriminant analysis employing infrared micro-spectroscopic data of entire yeast cells

This paper shows the ease of application and usefulness of mid-IR measurements for the investigation of orthogonal cell states on the example of the analysis of Pichia pastoris cells. A rapid method for the discrimination of entire yeast cells grown under carbon and nitrogen-limited conditions based on the direct acquisition of mid-IR spectra and partial least squares discriminant analysis (PLS-DA) is described. The obtained PLS-DA model was extensively validated employing two different validation strategies: (i) statistical validation employing a method based on permutation testing and (ii) external validation splitting the available data into two independent sub-sets. The Variable Importance in Projection scores of the PLS-DA model provided deeper insight into the differences between the two investigated states. Hence, we demonstrate the feasibility of a method which uses IR spectra from intact cells that may be employed in a second step as an in-line tool in process development and process control along Quality by Design principles.

Granulation behaviour of increasingly hydrophobic mixtures

Abstract Wet granulation is the formation of powder particle assemblies bound by a liquid binder. Granules are produced in order to prevent mixture segregation, improve flow characteristics and ease of powder handling. The liquid binder droplets penetrate into, and bind together, powder particles being agitated in a mixer. Real granulation formulations are complex and challenging. The difficulty of achieving controlled water based wet granulation of a formulation containing a hydrophobic component is one such challenge that is receiving increasing attention in academic and industrial research. The present work has focussed on the granulation of multicomponent batches, specifically mixtures of hydrophilic Granulac230 (lactose) and hydrophobic limestone, using different spray droplet sizes and impeller speeds. The granulation behaviour of the batches is assessed in terms of granule size and compositional distributions on increasing the hydrophobic content of the batch. Transitions in granulation behaviour are reported as the hydrophobic content of the batch is increased. The granulation behaviour is linked to static powder bed penetration times by assigning ‘compositional zones’, which could be used as a predictive tool in formulation and process development In addition, a new preferential nucleation and layering mechanism is proposed to describe the experimental data and better understand how the ‘compositional zones’ of granulation behaviour have arisen.

Uncertainty in techno-economic estimates of cellulosic ethanol production due to experimental measurement uncertainty

BackgroundCost-effective production of lignocellulosic biofuels remains a major financial and technical challenge at the industrial scale. A critical tool in biofuels process development is the techno-economic (TE) model, which calculates biofuel production costs using a process model and an economic model. The process model solves mass and energy balances for each unit, and the economic model estimates capital and operating costs from the process model based on economic assumptions. The process model inputs include experimental data on the feedstock composition and intermediate product yields for each unit. These experimental yield data are calculated from primary measurements. Uncertainty in these primary measurements is propagated to the calculated yields, to the process model, and ultimately to the economic model. Thus, outputs of the TE model have a minimum uncertainty associated with the uncertainty in the primary measurements.ResultsWe calculate the uncertainty in the Minimum Ethanol Selling Price (MESP) estimate for lignocellulosic ethanol production via a biochemical conversion process: dilute sulfuric acid pretreatment of corn stover followed by enzymatic hydrolysis and co-fermentation of the resulting sugars to ethanol. We perform a sensitivity analysis on the TE model and identify the feedstock composition and conversion yields from three unit operations (xylose from pretreatment, glucose from enzymatic hydrolysis, and ethanol from fermentation) as the most important variables. The uncertainty in the pretreatment xylose yield arises from multiple measurements, whereas the glucose and ethanol yields from enzymatic hydrolysis and fermentation, respectively, are dominated by a single measurement: the fraction of insoluble solids (fIS) in the biomass slurries.ConclusionsWe calculate a $0.15/gal uncertainty in MESP from the TE model due to uncertainties in primary measurements. This result sets a lower bound on the error bars of the TE model predictions. This analysis highlights the primary measurements that merit further development to reduce the uncertainty associated with their use in TE models. While we develop and apply this mathematical framework to a specific biorefinery scenario here, this analysis can be readily adapted to other types of biorefining processes and provides a general framework for propagating uncertainty due to analytical measurements through a TE model.

Transfer Molding Technology for Smart Power Electronics Modules: Materials and Processes

In recent years, within power electronics packaging, there has been a trend toward compact power electronics modules for automotive and industrial applications, where a smart integrated control unit for motor drives is replacing bulky substrates with discrete control logic and power electronics. Most recent modules combine control and power electronics, yielding maximum miniaturization. Transfer molding is the method of choice for cost-effective encapsulation of such modules due to robustness of the molded modules and moderate cost of packaging. But there are challenges with this type of package. Typically, these packages are asymmetric, and thus a substrate with single sided assembly is overmolded on the component side and the substrate backside is exposed, providing a heat path for optimized cooling. This asymmetric geometry is prone to yielding warped substrates, preventing optimum thermal contact to the heat sink and also putting thermomechanical stress on the encapsulated components, possibly reducing reliability. Such packages are truly heterogeneous, combining power ICs, wire bonds, SMDs, control ICs, substrate, and lead frame surfaces. As a result, the encapsulant used needs to adhere sufficiently to all surfaces present. Additionally, those packages need to operate at elevated temperatures for extended time periods, for example, at 150°C for 2000 h and more, so high thermal stability is of prime importance. Within this paper, a reference application is described integrating power and control logic inside a lead frame based molded package. Taking into account the challenges mentioned above, a detailed description of material selection for this module will be given, including material analysis, such as rheology, reactivity, and change in εr; and thermomechanical properties, in initial stage as f(t,T) and after media storage. Process development tools for module molding are used to ensure manufacturability and usability. Concluding rules for encapsulant material selection and package setup are provided.

Assessment of AMBR<sup>TM</sup> as a model for high-throughput cell culture process development strategy

The development and delivery of high quality therapeutic products necessitates the need for highthrough-put (HTP) process development tools. Traditionally, these works requires a combination of shake flask and small-scale stirred tank bioreactor (STR) which are labor and resource intensive and time-consuming. Here we demonstrate a strategy for rapid and robust cell culture process development by evaluating and implementing the use of a new HTP disposable micro bioreactor (MBR) called AMBRTM system (Advanced Microscale Bioreactor) that has the capabilities for automated sampling, feed addition, pH, dissolved oxygen (DO), gassing and agitation controls. In these studies the performance of two monoclonal antibody (MAb) producing cell lines (MAb1 and MAb2) was evaluated both in the AMBR system and 3-L STR. We demonstrated that cell culture performance (growth and viability, production titer and product quality) were similar in both vessel systems. Furthermore, process control and feed optimization were demonstrated in an additional cell line (MAb3) in the disposable MBR and its performance confirmed at STR scale. The results indicate that the AMBR system can be used to streamline the process development effort and facilitate a rapid and robust cell culture process development effort for MAb programs in a HTP manner.

Sustainability assessment of novel chemical processes at early stage: application to biobased processes

Chemical conversions have been a cornerstone of industrial revolution and societal progress. Continuing this progress in a resource constrained world poses a critical challenge which demands the development of innovative chemical processes to meet our energy and material needs in a sustainable way. This challenge forms the basis for this article. We report a method for quick preliminary assessment of chemical processes at the laboratory stage. The proposed method enables a review of chemical processes within a broader sustainability context. It is inspired by green chemistry principles, techno-economic analysis and some elements of environmental life-cycle assessment (LCA). This method evaluates a proposed chemical process against comparable existing processes using a multi-criteria approach that integrates various economic and environmental indicators. An effort has been made to incorporate quantitative and qualitative information about the processes while making the method transparent and easy to implement based on information available at an early stage in process development. The idea is to provide a data-based assessment tool for chemists and engineers to develop sustainable chemistry. This paper describes the method in detail and examines plausibility of the results. A biobased process for the production of but-1,3-diene has been analyzed using this method. This biobased process is compared with a conventional process for the production of but-1,3-diene from petroleum sources. The effects of uncertainty in the underlying model parameters and assumptions are also analyzed, along with the effect of system boundary selection on the assessment outcome. Analysis and testing of the method shows that it can be used as a valuable tool for sustainable process development.

Data Driven Modelling and Control of Batch Processes in the Pharmaceutical Industry

Regulatory changes in the pharmaceutical industry have recast the role of chemical engineering in the development of processes for the large-scale manufacture of Active Pharmaceutical Ingredients. Health authorities' expectations have increased regarding proven demonstration of adequate process performance, process robustness, and quality control across development and manufacturing scales. These expectations have substantially increased the demands of experimental data collection, process monitoring, and multivariate process understanding. To address these requirements, innovative implementation of established and emerging automation, modeling, data management, and process monitoring techniques have been increasingly added to the repertoire of process development tools in particular, those carried out in agitated stirred tanks. This article will introduce the application of these techniques across the range of process development (from early to late phase) challenges. In particular, this article describes the implementation of Data Exploration Analysis to results obtained with automated parallel experimentation for batch reaction characterization and automated batch crystallization processes integrated to population balance modeling. Collectively these examples of new strategies for the automation of experimental batch processes, data analysis, and modeling provide an overview of recent trends in pharmaceutical chemical process development.

Decision support tool for lean product and process development

Lean and concurrent engineering (CE) are widely acknowledged business process improvement strategies. These strategies can improve processes, reduce costs, and cut waste enabling organisations to remain competitive. Lean manufacturing offers an enterprise-wide methodology that improves reliability and flexibility while reducing lead-times and inventory carrying costs. Companies in manufacturing and service sectors are focusing on integrating lean manufacturing methodology with other applications, so that, all their systems and processes are aligned. This article proposes a CE framework based on application of information technology and object-oriented methodology for lean manufacturing. This approach expected to give manufacturing companies an extra edge in today's competitive market. A case study is presented in this article to demonstrate the effectiveness of CE framework in a machine tool manufacturing company. CE practice was adopted for new products, to align the systems and processes of the company. Product development lead-time was found to decrease by more than 50% compared to similar development projects carried out by the company. The need for rework was found to be negligible and the development cost was reduced considerably.

A methodology for qualifying industrial CFD: The Q 3 approach and the role of a protocol

Computational Fluid Dynamics (CFD) analysis has become a useful research and design instrument albeit with errors and uncertainties. Appropriate standards and protocols for increasing confidence and reliability need to be identified and applied. These requirements become more relevant as improvements in hardware, software and user competence increase. Advances in these sectors have led to the increased use of CFD codes in the applied research industry. In the discussion on the opportunities of applying Quality Assurance to Research and Development (R&D) activities, and on investigating methodologies (numerical or experimental), CFD results need to be qualified; rules, and procedures should be instated and followed. This paper proposes a methodological approach to qualify CFD, named Q 3 approach. The approach is based on three interdependent, but related, dimensions: software reliability, user knowledge and process control. Applying the quality of CFD analysis to industrial problems, while following the principles of Quality Assurance, is of particular concerned. The paper aims to focus on the dimensions of user knowledge and process control, which are often overlooked and are of great importance when determining the quality of CFD. The approach discussed here refers to a typical R&D department in a Small or Medium Enterprise (SME) where a CFD code is used as a support tool for process or product development, optimisation and/or innovation. Within the three dimensions, software reliability immediately refers to Verification and Validation (V&V) procedures, a well known issue for those who are involved in code and model development. User knowledge and process control play a vital role in achieving accurate results and cannot be neglected, as per the principles of Quality Assurance. Therefore it is vital to provide continuous CFD analyst training to industrial professionals. Applying a CFD cycled process, which refers to the cyclic logic proposed in Quality Assurance, may face some constrains in specific industrial scenarios. A protocol will serve as an instrument of process control. Within the protocol, the phases of Calculation, Verification and Validation as well as User Defined Model Verification represent critical elements, and are specifically detailed and structured in accordance with the RADAR logic (Results, Approach, Deploy, Assess, Refine) used for Quality Assurance Management. In the case a specific study does not rely on experimental data, an alternative is proposed. The protocol is designed as a single tool that responds to both the need for qualified CFD and the requirement of internal or external audit of Quality Assurance (as far as calculations are concerned), thus increasing cost and personnel efficiency.

Model simulation and experimental verification of a cation-exchange IgG capture step in batch and continuous chromatography

The cation-exchange capture step of a monoclonal antibody (mAb) purification process using single column batch and multicolumn continuous chromatography (MCSGP) was modeled with a lumped kinetic model. Model parameters were experimentally determined under analytical and preparative conditions: porosities, retention factors and mass transfer parameters of purified mAb were obtained through a systematic procedure based on retention time measurements. The saturation capacity was determined through peak fitting assuming a Langmuir-type adsorption isotherm. The model was validated using linear batch gradient elutions. In addition, the model was used to simulate the start-up, cyclic steady state and shut down behavior of the continuous capture process (MCSGP) and to predict performance parameters. The obtained results were validated by comparison with suitable experiments using an industrial cell culture supernatant. Although the model was not capable of delivering quantitative information of the product purity, it proved high accuracy in the prediction of product concentrations and yield with an error of less than 6%, making it a very useful tool in process development.

Plasmid DNA production for pharmaceutical use: Role of specific growth rate and impact on process design

Application of plasmid DNA as pharmaceutical to be used in gene therapy and vaccination has been investigated intensively in recent years. To be able to provide sufficient material that is in accordance with quality of pharmaceutical grade it is mandatory to gain comprehensive process knowledge which is even requested by regulatory agencies. Regarding plasmid DNA production the specific growth rate has been identified as one of the key parameters. The reduction of specific growth rate results in an increase of plasmid DNA formation. However, quantitative explanations that allow for efficient process development and design are still missing. The presented study proposes a model that clearly demonstrates the relationship between specific growth rate and plasmid formation due to identification of the specific plasmid production rate as relevant key parameter. In addition the model is proved to serve as a useful tool in process development and design.

Transfer molding technology for smart power electronics modules - materials and processes

During the last years within power electronics packaging a trend towards compact power electronics modules for automotive and industrial applications could be observed, where a smart integrated control unit for motor drives is replacing bulky substrates with discrete control logic and power electronics. Most recent modules combine control and power electronics yielding maximum miniaturization. Transfer molding is the method of choice for cost effective encapsulation of such modules due to robustness of the molded modules and moderate cost of packaging. But there are challenges with this type of package: Typically those packages are asymmetric, a substrate with single sided assembly is overmolded on the component side and the substrate backside is exposed providing a heat path for optimized cooling. This asymmetric geometry is prone to yield warped substrates, preventing optimum thermal contact to the heatsink and also putting thermomechanical stress on the encapsulated components, possibly reducing reliability. Such packages being truly heterogeneous, combining powerICs, wire bonds, SMDs, controlICs, substrate and leadframe surfaces, the encapsulant used needs to adhere sufficiently to all surfaces present. Additionally those packages need to operate at elevated temperatures for increased times, e.g. operate at 150 °C for 2000 h and more, so high thermal stability is of ample importance. Within this paper a reference application is described, integrating power and control logic inside a leadframe based molded package. Taking into account the challenges mentioned above, a detailed description of material selection for this module will be given, including material analysis as rheology, reactivity, change in ɛr and thermomechanical properties as f(t,T) and of media storage. Process development tools for module molding are used to ensure manufacturability and useability. Concluding rules for encapsulant material selection and package setup are provided.

Testing coatings: An overview

Testing is an important tool for product and process development, for qualification as a supplier, and for monitoring the process and product over time. The test results are a demonstration of competence of personnel in production and in the lab. It should be a point of pride that the testing is done well and is useful to personnel in the lab, production, and sales.

Crucible-free pulling of germanium crystals

Commonly, germanium crystals are grown after the Czochralski (CZ) method. The crucible-free pedestal and floating zone (FZ) methods, which are widely used for silicon growth, are hardly known to be investigated for germanium. The germanium melt is more than twice as dense as liquid silicon, which could destabilize a floating zone. Additionally, the lower melting point and the related lower radiative heat loss is shown to reduce the stability especially of the FZ process with the consequence of a screw-like crystal growth. We found that the lower heat radiation of Ge can be compensated by the increased convective cooling of a helium atmosphere instead of the argon ambient. Under these conditions, the screw-like growth could be avoided. Unfortunately, the helium cooling deteriorates the melting behavior of the feed rod. Spikes appear along the open melt front, which touch on the induction coil. In order to improve the melting behavior, we used a lamp as a second energy source as well as a mixture of Ar and He. With this, we found a final solution for growing stable crystals from germanium by using both gases in different parts of the furnace. The experimental work is accompanied by the simulation of the stationary temperature field. The commercially available software FEMAG-FZ is used for axisymmetric calculations. Another tool for process development is the lateral photo-voltage scanning (LPS), which can determine the shape of the solid–liquid phase boundary by analyzing the growth striations in a lateral cut of a grown crystal. In addition to improvements of the process, these measurements can be compared with the calculated results and, hence, conduce to validate the calculation.

Physiological characterisation of S. cerevisiae cells expressing the B. subtilis alanine dehydrogenase gene

This paper shows the ease of application and usefulness of mid-IR measurements for the investigation of orthogonal cell states on the example of the analysis of Pichia pastoris cells. A rapid method for the discrimination of entire yeast cells grown under carbon and nitrogen-limited conditions based on the direct acquisition of mid-IR spectra and partial least squares discriminant analysis (PLS-DA) is described. The obtained PLS-DA model was extensively validated employing two different validation strategies: (i) statistical validation employing a method based on permutation testing and (ii) external validation splitting the available data into two independent sub-sets. The Variable Importance in Projection scores of the PLS-DA model provided deeper insight into the differences between the two investigated states. Hence, we demonstrate the feasibility of a method which uses IR spectra from intact cells that may be employed in a second step as an in-line tool in process development and process control along Quality by Design principles.