他们的问题，
和我们的解法。Their problems,
and how we solved them.

面临的问题The challenge

生产订单靠 Excel 在车间、仓库、质检三方之间传递，版本经常对不上；客户要求提供产品全过程追溯时，需要人工翻找两个月的纸质记录。Production orders moved between the shop floor, warehouse, and QA team in Excel, and version mismatches were routine. When a customer requested full product traceability, staff had to dig through two months of paper records by hand.

我们的方案Our approach

• —梳理 14 类工单模板与 23 个关键流转节点，替换全部线下 ExcelMapped 14 work order templates and 23 key handoff points, replacing all offline Excel workflows
• —搭建统一工序编码规则，实现批次级全链路追溯Established a unified process coding standard for end-to-end batch-level traceability
• —对接已有 ERP 与财务系统，避免重复录入Integrated with the existing ERP and finance systems to eliminate duplicate data entry
• —车间大屏实时显示在制工单进度与异常件Rolled out shop floor displays showing live work order progress and exception items

面临的问题The challenge

门店系统与电商后台各自为政，同一顾客在两边是两个身份；活动效果只能看总营收，无法判断哪一次活动真正带来了复购。Store systems and the e-commerce back office operated independently, so the same customer had two separate identities. Campaign performance could only be measured by total revenue—impossible to tell which campaign actually drove repeat purchases.

我们的方案Our approach

• —建立统一会员主数据模型，完成存量会员去重与合并Built a unified customer master data model and de-duplicated and merged existing member records
• —打通门店 POS、电商与小程序三端的会员与储值数据Connected membership and stored-value data across in-store POS, e-commerce, and the mini program
• —重构活动分析指标体系，支持单次活动的 ROI 归因Redesigned campaign analytics to attribute ROI to individual campaigns
• —上线门店与区域两级经营看板Delivered store-level and regional performance dashboards

面临的问题The challenge

货源信息靠微信群和电话协调，运力调度全凭调度员经验；出了问题追溯不到具体批次，消费者信任度持续走低。Supply information was coordinated over WeChat groups and phone calls, and dispatching relied entirely on individual dispatchers' experience. When problems arose, there was no way to trace a specific batch, and consumer trust kept declining.

我们的方案Our approach

• —设计产地分级、质检、入库、发货五段式溯源编码Designed a five-stage traceability coding scheme covering origin grading, quality inspection, warehousing, and shipping
• —开发基于订单时效与车辆位置的运力调度看板Built a dispatch dashboard driven by order deadlines and real-time vehicle locations
• —搭建面向合作社的轻量化移动端录入工具Developed a lightweight mobile data entry tool for cooperatives
• —对接监管平台完成数据上报Integrated with regulatory platforms for automated data reporting

面临的问题The challenge

运单状态靠电话和微信更新，货损责任无法追溯；财务每月对账要花两周，仍有争议款项说不清。Waybill status was updated via phone calls and WeChat, leaving cargo damage claims untraceable. Monthly reconciliation took finance two weeks and still left disputed amounts unresolved.

我们的方案Our approach

• —统一运单状态机与异常处理流程Standardized the waybill status model and exception handling workflow
• —对接 GPS 与车载终端实现位置回传Integrated GPS and in-vehicle terminals to report real-time location data
• —建立自动计费规则引擎与对账报表Built an automated freight billing rules engine and reconciliation reports
• —异常索赔线上化流转Moved damage claims to an online approval workflow

面临的问题The challenge

四个系统各有一份学生数据，转专业、评奖评优等流程要在多系统之间人工核对，每年招生季行政人员要加班三周。Each of the four systems held its own copy of student data, so processes such as major transfers and scholarship decisions required manual cross-checking between systems. Every admissions season, admin staff worked three weeks of overtime.

我们的方案Our approach

• —建立学生主数据中心并完成四个系统的数据归一Built a student master data hub and normalized data across the four systems
• —重做选课、成绩、评奖评优等 9 条核心业务流程Redesigned 9 core workflows, including course selection, grading, and scholarships
• —开放学生端与教师端移动应用Launched mobile apps for students and faculty
• —提供历史数据迁移与双轨并行方案Delivered historical data migration and a parallel-run cutover plan

面临的问题The challenge

全部预约依赖电话，前台在高峰期漏接与错约频发；术后随访靠纸质记录，科研数据难以沉淀。All appointments were booked by phone, and front desk staff frequently missed calls or booked the wrong slot during peak hours. Post-surgery follow-ups were recorded on paper, making it hard to build research datasets.

我们的方案Our approach

• —开发分诊规则引擎，按科室与医生维度分配号源Built a triage rules engine to allocate appointment slots by department and physician
• —上线患者端小程序与自助取号Launched a patient mini program with self-service check-in
• —设计结构化随访表单并回流临床数据Designed structured follow-up forms that feed data back into clinical records
• —建立院区经营与医疗质量双看板Built dual dashboards for hospital operations and care quality

面临的问题The challenge

纯人工审批积压严重，且不同审批员口径不一致；风控规则散落在多份 Excel 中，无法审计。Manual-only approval created a serious backlog, and reviewers applied inconsistent criteria. Risk rules were scattered across multiple Excel files and could not be audited.

我们的方案Our approach

• —梳理并固化 68 条审批规则，全部版本化可追溯Documented and formalized 68 approval rules, all version-controlled and auditable
• —开发规则引擎与审批工作台Built a rules engine and reviewer workbench
• —建立反欺诈特征库与人工复核队列Created an anti-fraud feature library and manual review queue
• —接入征信与工商数据源做交叉验证Integrated credit bureau and business registry data for cross-validation

面临的问题The challenge

不同部门各有一套申报入口与材料要求，企业反复提交同一份资料；政策文件分散，看得见却用不上。Each department had its own application portal and document requirements, forcing companies to resubmit the same materials repeatedly. Policies were scattered across documents—visible but hard to actually use.

我们的方案Our approach

• —梳理 26 项政务事项的统一材料清单Standardized document checklists for 26 government service items
• —建设政策智能匹配引擎，按企业画像推送可申报政策Built a smart policy matching engine that recommends applicable policies based on company profiles
• —开发企业诉求工单与闭环评价系统Developed a ticketing system for business requests with closed-loop feedback
• —完成与上级政务平台的接口对接Completed API integration with higher-level government service platforms