A reliability-oriented placement algorithm for reconfigurable digital microfluidic biochips using 3-D deferred decision making technique

In recent studies, digital microfluidic biochips (DMFBs) have been a promising solution for lab-on-a-chip and bio-assay experiments because of their flexible application and low fabrication cost. However, the reliability problem is an imperative issue to guarantee the valid function of DMFBs. The reliability of DMFBs decreases when electrodes are excessively actuated, preventing droplets on DMFBs controlled successfully. Because the placement for bio-assays in DMFBs is a key step in generating corresponding actuating signals, the reliability of DMFBs must be considered during biochip placement to avoid excessive actuation. Although researchers have proposed several DMFB placement algorithms, they have failed to consider the reliability issue. In addition, previous algorithms were all based on the simulated-annealing (SA) method, which is time consuming and does not guarantee to obtain an optimal solution. This paper proposes the first reliability-oriented non-SA placement algorithm for DMFBs. This approach considers the reliability problem during placement, and uses the 3-D deferred decision making (3D-DDM) technique to enumerate only possible placement solutions. Large-scale DMFB placement can be synthesized efficiently by partitioning the operation sequential graph of bioassays. Experimental results demonstrate that the proposed technique can achieve reliability-oriented placement for DMFBs without excessive actuation in each electrode, while optimizing bioassay completion time.