Requirements specification, automated implementation

Sample dilution is one of the most labor-intensive tasks in running an LBA. Add standard curve and quality control dilution (which these systems can easily do) [8], and it becomes over half the effort to run an assay. Depending on the specific practices of a laboratory, samples can be diluted in duplicate, triplicate, at four serial dilutions, at eight serial dilutions, with one or more sample diluents, at different dilutions for each sample, and from different size tubes (e.g., clinical versus nonclinical). Similarly, standard curve and control dilutions can be performed in a variety of ways. For these reasons, it can be challenging to implement an automated dilutor to meet all of these requirements. 

In droplet-based microfluidics, these reaction vessels are formed by droplets of a dispersed phase, which are embedded into a continuous phase. Both liquid phases are immiscible. A huge amount of such droplet reactors can be generated, transported, controlled, and processed in parallel in a droplet-based lab-on-a-chip device. These devices can be characterized as application specific microfiuidic networks that implement and automate a conventional laboratory workflow in a microfluidic chip device or system. They are built up by appropriately intercoimecting microfluidic operation units, which provide the required laboratory operations at the microscale. Consequently, for each conventional laboratory operation, its microscale counterpart is required. 

There are several limitations that need to be addressed. The proposed approach allows for automated development of strucmre of simulation models while information models driven development of control mechanisms is supported to a limited extent. Improving implementation of control mechanisms is a direction of further research. The approach can only be efficient if high reusability is achieved. Therefore, a precise specification of model transformation and generation mechanisms is required. 

In this paper we present a formal approach allowing to automate the process of analyzing the compliance of the fault detection and mitigation capabilities of an implementation with the safety specification. To achieve this, the safety requirements need to be stated in a formal way. There are multiple formalization approaches for safety concepts, starting from the very theoretical function-structure-models [5] or the notion of processes and channels [12], which are both not suitable for production use, since they require a special structural model. The HipHops approach [15,14] can be applied to development models like Simulink, but lacks some important features such as timing support. In this paper we are focusing on the concept of safety contracts [13]. 

---