Judicial Case Management System

JCMS shows that prioritisation, duration estimation, hearing scheduling and citizen notification can be combined in one system that keeps a human in charge and states its uncertainty. On synthetic data, a fixed-weight P-Score reduces the wait of undertrial cases from 168 to 4 days, and the plain greedy scheduler starves 48 of 750 old cases; a protected 10% quota removes that starvation without changing the formula. The duration models are only slightly better than a baseline, which the noise ceiling explains. Future work is to validate the method on real data with court partners, to test an age term based on time since the last hearing, to let courts choose the protected share per week, and to complete the language review and screen-reader testing.

[1]               National Judicial Data Grid, eCourts Services, Department of Justice, Government of India. [Online]. Available: https://njdg.ecourts.gov.in/ (accessed Oct. 27, 2026). [2]               "Jammu Kashmir High Court pendency rises to 44,551 cases; subordinate courts cross 3.65 lakh, Centre tells Lok Sabha," Kashmir Life, Jul. 2026. [Online]. Available: https://kashmirlife.net/ (accessed Oct. 27, 2026). [3]               National Crime Records Bureau, Prison Statistics India 2022. New Delhi, India: Ministry of Home Affairs, Government of India, 2023. [4]               Law Commission of India, "Arrears and backlog: Creating additional judicial (wo)manpower," Rep. No. 245, Jul. 2014. [5]               J. Angwin, J. Larson, S. Mattu, and L. Kirchner, "Machine bias," ProPublica, May 23, 2016. [6]               N. Aletras, D. Tsarapatsanis, D. Preoţiuc-Pietro, and V. Lampos, "Predicting judicial decisions of the European Court of Human Rights: A natural language processing perspective," PeerJ Comput. Sci., vol. 2, p. e93, 2016. [7]               D. M. Katz, M. J. Bommarito II, and J. Blackman, "A general approach for predicting the behavior of the Supreme Court of the United States," PLoS ONE, vol. 12, no. 4, p. e0174698, 2017. [8]               S. M. Lundberg and S.-I. Lee, "A unified approach to interpreting model predictions," in Proc. Adv. Neural Inf. Process. Syst. (NIPS), vol. 30, 2017, pp. 4765–4774. [9]               S. M. Lundberg et al., "From local explanations to global understanding with explainable AI for trees," Nat. Mach. Intell., vol. 2, no. 1, pp. 56–67, Jan. 2020. [10]             G. Ke et al., "LightGBM: A highly efficient gradient boosting decision tree," in Proc. Adv. Neural Inf. Process. Syst. (NIPS), vol. 30, 2017, pp. 3146–3154.