Managing The Uncertainty In Product Design And Development

New products are critical for the survival and success of enterprises.Generally,there are two major tasks for the design and development of new products.That is,identifying a proper product scheme and then turning the scheme into a deliverable product.Due to the innovation of product design and development,uncertainty challenges the practitioners.Therefore,this dissertation implement several researches on the uncertainty in product design and development.For identifying a proper product scheme,a major uncertainty lies in the imprecise characterization of customer preferences.This research explores how to deal with the uncertainty of customer preference in the share-of-choice problem(i.e.maximizing the percentage of buyers).We propose a distributionally robust optimization model that maximizes the worst case of the proportion of buyers over a Wasserstein uncertain set.The model offers the robustness of the designed product to customer preference uncertainty.Our model contributes both analytically and practically.From analytical aspect,we show that the key to obtain the robustness is to classify the buyers into loyal and disloyal ones according to the consumer surplus,and give priority to the loyal buyers.This guarantees a lower bound of buyers’ welfare(the summation of consumer surplus of buyers)in the meanwhile of share-of-choice optimization.For product line design,we also prove that higher product diversity can better deal with customer preference uncertainty in product line design.From practical aspect,our model provides a more powerful tool for product design.Through numerical experiments based on simulation and empirical data,we verify that the products designed by the distributionally robust optimization method result higher percentage of buyers and buyer loyalty,more stable market performance,and more accurate prediction on the percentage of buyers than those designed by the previous methods under the real customer preference distribution.To turn the select product scheme into a deliverable product,an important uncertainty is in the form of iteration.The effect of iteration on product development process may be either positive or negative.Therefore,how to effectively take advantages of the positive effect of iteration and avoid its negative effect plays an important role in improving the efficiency of product development.We investigate how to coordinate between faster and fewer iterations to minimize the lead time of product development.Through theoretical and numerical analysis,we reveal the impact of the characteristics,such as acceptance efficiency,rework risk,number of stages,location of each stage,on the iterative strategy of product development.In addition,we validate the effectiveness of the proposed model through empirical data.In the case study,a reasonable balance between the number and duration of iterations results in a 11.25% reduction of product development lead time.At the end of the dissertation,we conclude the above researches and provide several directions of future research.