Next-Generation Algorithm Training Research to Expedite AI Adoption and Accelerate Pandemic Resilience in Trade

自中世纪以来，货运代理代表a)空运、海运或陆运托运人以及b)承运人组织运输，以便将实物货物从A运往B。传统的劳动密集型纸质运输方式导致周转时间长、几乎没有正式的知识获取和高操作风险。CargoLogik承认“目前的流程是一个巨大的时间和开销杀手……仍然在很大程度上是人工的……这会导致货运代理损失时间，以及笨拙、糟糕的客户体验。”世界经济论坛强调，新冠肺炎暴露了物流实体和人工数据处理操作中的这些系统性弱点，主要是电子邮件和贸易文件驱动的工作流。这场流行病加剧了现有的低效和缺乏审查，每年给财富500强公司造成810亿美元的不必要供应链成本(摩根大通2017年贸易展望)。尽管世界经济论坛关于数字化创新的建议提供了补救措施，以提高迫切需要的业务弹性，但有效采用仍存在挑战。物流业采用人工智能的速度非常缓慢，目前只有12%的组织利用人工智能(MHI行业报告，2020)。缺乏人工智能应用的一个主要原因是物流部门拥有不断变化的、非标准的和复杂的信息，这给算法的可扩展性带来了巨大的挑战。该项目旨在通过开发一个无代码的、端到端的自动化算法训练管道来解决人工智能的可达性和采用率问题。该工具将内置到我们现有的物流机器学习操作工作流平台中，管道将在后台运行，自动重新培训我们的数据提取算法以适应客户不断变化的电子邮件和文档内容，从而通过轻松扩展和效率鼓励快速采用。

Since medieval times, freight forwarders have organized transportation on behalf of a) shippers, by air, ocean or land, and b) carriers, to get physical goods from A to B. The traditional labour-intensive paper driven approaches result in long turn-around times, little formal knowledge capture and high operational risk. CargoLogik acknowledges the "current processes are an immense time and overhead killer... still largely manual ...which result in lost time for freight forwarders and a clunky, poor customer experience." The World Economic Forum (WEF) highlighted that COVID-19 exposed these systemic weaknesses in Logistics' physical and manual data-processing operations, primarily email and trade-document driven workflows. The pandemic compounds existing inefficiencies and lack of scrutiny, costing Fortune 500 companies $81 billion of unnecessary supply chain costs each year (JPMorgan 2017 Trade Outlook).Whilst the WEF recommendation of digitisation innovations offer remedies to improve desperately needed business resiliency, there are challenges to effective adoption. The Logistics industry has been notably slow to adopt AI, only 12% of organisations currently leverage AI (MHI Industry report, 2020). One major reason for the lack of AI adoption is that the Logistics sector has ever-changing, non-standard and complex information, which poses massive algorithm scalability challenges.This project aims to address AI accessibility and adoption by developing a no-code, end-to-end automated algorithm-training pipeline. The tool will be in-built into our existing logistics' machine learning operations workflow platform and the pipeline will operate in the background, automatically re-training our data-extraction algorithms to customers' evolving email and document content, thus encouraging rapid adoption via scaling ease and efficiency.