Analyzing the Relationship between the Systems Decisions Process and Artificial Intelligence: A Machine Vision Case Study

Authors: Mitchell J. McHugh, John J. Case

Abstract:

Systems engineering is a holistic discipline that seeks to organize and optimize complex, interdisciplinary systems. With the growth of artificial intelligence, systems engineers must face the challenge of leveraging artificial intelligence systems to solve complex problems. This paper analyzes the integration of systems engineering and artificial intelligence and discusses how artificial intelligence systems embody the systems decision process (SDP). The SDP is a four-stage problem-solving framework that outlines how systems engineers can design and implement solutions using value-focused thinking. This paper argues that artificial intelligence models can replicate the SDP, thus validating its flexible, value-focused foundation. The authors demonstrate this by developing a machine vision mobile application that can classify weapons to augment the decision-making role of an Army subject matter expert. This practical application was an end-to-end design challenge that highlights how artificial intelligence systems embody systems engineering principles. The impact of this research demonstrates that the SDP is a dynamic tool that systems engineers should leverage when incorporating artificial intelligence within the systems that they develop.

Keywords: Computer vision, machine learning, mobile application, systems engineering, systems decision process.

References:

[1] Parnell, G. S., Driscoll, P. J., & Henderson, D. H. (2011). Decision Making in Systems Engineering and Management. Hoboken: John Wiley & Sons, Inc..
[2] Petrotta, M., Sterling Heights, M. I., & Peterson, T. (2019, July). Implementing Augmented Intelligence in Systems Engineering. In INCOSE International Symposium (Vol. 29, No. 1, pp. 543-543).
[3] Sommerville, I. (2002). Artificial Intelligence and Systems Engineering. Lancaster University.
[4] Sessions, V., & Marco Valtorta. (2006). The Effects of Data Quality on Machine Learning Algorithms. ICIQ.
[5] Alding, O. (2018). Mobile Object Detection using TensorFlow Lite and Transfer Learning. Stockholm, Sweden: Digitala Vetenskapliga Arkivet.
[6] Cheng, Y., Wang, D., Zhou, P., & Zhang, T. (2017). A Survey of Model Compression and Acceleration for Deep Neural Networks. IEEE Signal Processing Magazine Sessions, V., & Marco Valtorta. (2006). The Effects of Data Quality on Machine Learning Algorithms. ICIQ.
[7] Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., & Chen, L.-C. (2018). MobileNetV2: Inverted Residuals and Linear Bottlenecks. The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 4510-4520.
[8] TensorFlow. (n.d.). TensorFlow Lite image classification Android example application. Retrieved from GitHub: https://github.com/tensorflow/examples/tree/master/lite/examples/image_classification/android