数据模板：应付账款发票处理

SAP ECC

应付账款发票处理属性 (22) 应付账款发票处理活动 (13) 提取指南 (4)

您的应付账款发票处理数据模板

这份全面的数据模板提供了有效分析“应付账款发票处理”所需的关键信息。它概述了需要收集的重要数据属性、要追踪的核心流程活动，以及直接从SAP ECC系统提取这些数据的实用指导。使用此模板可确保您捕获所有必要的细节，以进行全面而富有洞察力的流程分析。

建议收集的属性
需要跟踪的关键活动
提取指南

不熟悉事件日志？了解如何创建流程挖掘事件日志.

下载模板

应付账款发票处理属性

这些是建议您在事件日志中纳入的数据字段，以便全面分析您的应付发票处理。

3 必填 9 推荐 10 可选

	名称	描述
	事件时间 EventTime	特定活动或事件发生的精确日期和时间。
描述事件时间是流程活动的时间戳，对于事件按时间顺序排列和基于时间分析至关重要。此属性用于计算周期、处理和等待时间等关键绩效指标，识别瓶颈，并检查 SLA 或付款条款合规性。为何重要该timestamp对计算所有基于持续时间的指标（包括周期时间和瓶颈）至关重要，是绩效分析的基础。获取方式源自 SAP 表，例如 BKPF（CPUDT/CPUTM 用于创建）或 CDHDR（UDATE/UTIME 用于更改）。示例 2023-10-01T10:00:00Z2023-10-02T14:35:10Z2023-10-15T09:12:00Z
	发票 InvoiceNumber	每份发票凭证的唯一标识符，作为追踪其从接收到付款整个流程的主要案例标识符。
描述发票号码是一个组合键，在SAP系统中通常由公司代码 (BUKRS)、凭证号码 (BELNR) 和会计年度 (GJAHR) 组合而成。这种独特的组合确保了系统中的每张发票都能被唯一识别。在Process Mining中，该属性是核心属性。它将创建、过账、审批和付款等所有相关事件和活动关联起来，形成一个单一、连贯的流程实例。依据发票进行流程分析，可以获得发票生命周期的完整端到端视图，这对于计算周期时间、识别变体和理解返工循环至关重要。为何重要它是连接所有发票相关活动的必要 Case ID，从而能够对每张发票的应付账款流程进行完整的端到端分析。获取方式由SAP表BKPF（会计凭证抬头）中的BUKRS、BELNR和GJAHR字段构建。示例 1000-1900000001-20232000-5100000055-20231000-1900000042-2024
	活动 Activity	指发票处理生命周期中发生的特定业务步骤或事件的名称。
描述此属性描述了应付账款流程中的单一任务或状态变更，例如“发票暂存”、“发票待审批”或“发票已清账”。这些活动是流程图的基石，通常从SAP内部的更改日志、状态字段或工作流日志中提取。分析这些活动的顺序和持续时间是Process Mining的核心。它有助于可视化流程流、识别步骤间的瓶颈、衡量返工循环的频率，并将实际流程执行与标准操作程序进行比较。活动粒度的定义直接影响所能获得的洞察深度。为何重要活动是流程图的核心骨架，可实现流程可视化、识别瓶颈并分析偏差。获取方式根据发票凭证的变更，从交易代码（BKPF-TCODE）、工作流日志（SWW_WI2OBJ）或变更日志（CDHDR/CDPOS表）推导。示例发票凭证已创建发票已审批发票已通过付款清算设置付款冻结
	付款冻结原因 PaymentBlockReason	指示发票被冻结付款原因的代码。
描述当发票存在差异或正在审核时，可以对其进行付款冻结，以防止其被纳入付款批次。此属性提供了冻结的具体原因，例如数量不匹配或价格差异。这些信息对于异常分析至关重要。它有助于对付款延迟和返工的根本原因进行分类，并支持“异常和差异分析”dashboard。深入了解最常见的冻结原因，能促使企业进行有针对性的流程改进，从而减少异常情况的发生。为何重要揭示付款延迟的根本原因，从而实现有针对性的分析，以减少异常、返工并缩短发票处理时间。获取方式源自 SAP 表 BSEG（会计凭证项目）的 ZLSPR 字段。示例 RIA
	付款条款 PaymentTerms	定义付款条件的编码，例如到期日和潜在折扣。
描述付款条款是 SAP 中预定义的规则，用于确定发票到期日和现金折扣的计算方式。它们被分配到供应商主数据中，也可以在单个发票上指定。此属性对于“付款合规性与折扣”仪表板和“提前付款折扣获取率”关键绩效指标（KPI）至关重要。通过分析付款条款，公司可以识别获取折扣的机会，了解付款计划的财务影响，并确保合规性。为何重要定义了付款到期日和折扣的规则，这对于财务优化和合规监控至关重要。获取方式源自 SAP 表 BSEG（会计凭证项目）的 ZTERM 字段。示例 Z0010001NT30
	供应商名称 VendorName	提交发票的供应商名称。
描述此属性包含供应商的正式名称。在SAP中，供应商主数据（表LFA1）通过存储在凭证抬头或行项目中的供应商编号（LIFNR）关联到发票凭证。按供应商进行分析对于“供应商付款绩效”dashboard至关重要。它有助于追踪发票处理时间、付款及时性以及特定供应商的异常率。这些洞察有助于管理供应商关系、识别发票质量持续不佳的问题供应商，并协商更好的条款。为何重要支持针对供应商进行绩效分析，有助于管理供应商关系并识别与特定供应商相关的问题。获取方式源自 SAP 表 LFA1 的 NAME1 字段，通过在 BKPF 或 BSEG 中找到的供应商编号 (LIFNR) 进行关联。示例环球办公用品有限公司科技解决方案有限责任公司可靠物流公司
	公司代码 CompanyCode	在 SAP 组织中代表法律实体或公司的唯一键。
描述公司代码是SAP财务模块中的一个基本组织单位，代表一个独立的会计核算单位。包括发票在内的所有财务交易都会过账到特定的公司代码下。该属性对于按法人实体划分流程分析至关重要。它使得比较不同业务部门的流程绩效、合规率和效率成为可能，这对于拥有多个子公司的大型跨国公司而言尤其重要。为何重要支持按法人实体细分流程分析，便于比较不同业务单元的绩效表现。获取方式源自 SAP 表 BKPF（会计凭证抬头）的 BUKRS 字段。示例 10002100US01
	发票到期日 InvoiceDueDate	根据付款条款计算出的发票到期付款日期。
描述发票到期日是指向供应商付款的最后期限，确保符合双方约定的条款。在SAP系统中，这通常不是一个单独存储的字段，而是根据凭证的基准日期 (BSEG-ZFBDT) 和与发票关联的付款条款 (BSEG-ZTERM) 计算得出的。此日期对于“付款合规与折扣”dashboard至关重要。它用于计算逾期付款率等KPI，并识别有逾期付款风险的发票。依据此日期进行分析，有助于优化工作优先级并改善供应商关系。为何重要这个日期对监控付款合规性、计算逾期KPI以及争取早期付款折扣至关重要。获取方式基于到期日计算基准日期（BSEG-ZFBDT）和付款条件码（BSEG-ZTERM）推导得出。示例 2023-10-312023-11-152024-01-10
	发票总金额 InvoiceGrossAmount	发票的总金额，包含税费及其他费用，以原始凭证币种表示。
描述此属性表示发票的总财务价值。它是理解正在处理发票的财务影响和特征的关键数据。在SAP中，此金额可在各种表中找到，具体取决于凭证类型，例如财务凭证的BSEG表。在分析中，发票金额用于优先处理高价值发票，理解AP（应付账款）流程的财务吞吐量，并进行细分分析。例如，可以分析高价值发票是否具有更长的周期时间或遵循不同的审批路径。它也是各项财务KPI的基础数据。为何重要提供流程的财务背景，从而支持基于价值的分析、高价值发票的优先处理以及财务KPI的计算。获取方式源自 SAP 行项目表，如 BSEG 字段 WRBTR（以单据货币计的金额）。可能需要计算以将行项目汇总到表头级别。示例 1500.0012550.75980.50
	审批人 Approver	负责审批发票付款的用户或角色。
描述此属性识别执行“发票已审批”活动的人员。在SAP中，这通常通过workflow系统管理，审批人的ID会被记录在workflow历史中。通过审批人进行分析，是“发票审批瓶颈分析”dashboard的核心。它有助于衡量个人或团队的审批时间，识别持续缓慢的审批人，并了解审批层级内的任务分配情况。这些数据对于优化审批workflow至关重要，因为审批workflow往往是主要瓶颈。为何重要识别负责审批的个人，从而能够按个人或团队分析审批周期时间和瓶颈。获取方式通常通过将workflow项目链接到发票对象并识别审批步骤的用户，在SAP业务workflow表（例如SWW_WI2OBJ, SWWLOG）中找到。示例 DMARTINLCHENFINMAN_ROLE
	流程处理用户 ProcessorUser	执行此活动（例如过账或更改发票）的员工的SAP用户名。
描述此属性用于识别负责执行特定流程步骤的个人用户。在SAP中，这可能是创建凭证的用户（例如BKPF-USNAM），也可能是进行了特定更改的用户（例如CDHDR-USERNAME）。通过处理员用户进行分析，是理解团队绩效和工作量分配的关键。它通过追踪每位用户的发票量和平均活动时间，帮助构建“应付账款处理员工作量分布”dashboard。这可以揭示培训需求，识别优秀员工，并支持公平的任务分配。为何重要有助于进行工作量分析、用户间绩效比较，并识别培训机会或资源配置不均衡的情况。获取方式源自 SAP 表 CDHDR 的 USERNAME 字段（用于更改）或 BKPF 的 USNAM 字段（用于单据录入）。示例 AJONESSMITHBCWILLIAMS
	采购订单号 PurchaseOrderNumber	发票所依据的采购订单标识符。
描述采购订单 (PO) 号将发票与采购流程关联起来。对于基于采购订单的发票，这种关联对于采购订单、收货和发票之间的三方匹配至关重要。此属性是“采购订单/收货匹配绩效”dashboard的关键。分析采购订单的存在和匹配情况，有助于区分不同类型的发票（带采购订单的发票与不带采购订单的发票），并衡量匹配流程的效率。在此领域出现的延误是流程瓶颈的常见来源。为何重要将发票与采购流程关联起来，有助于分析有无采购订单发票的区别，并评估三方匹配流程的效率。获取方式源自 SAP 表 BSEG（用于 FI 发票）或 RSEG（用于 MM 发票）的 EBELN 字段。示例 450001712345000175894500018330
	上次数据更新 LastDataUpdate	指示此记录数据上次从源系统刷新时间的D时间戳。
描述此属性标记从SAP ECC进行的最新数据提取或更新的日期和时间。它不属于交易数据本身，而是数据导入（data ingestion）过程中添加的metadata。在分析中的主要用途是提供所查看数据的时效性背景。这对于dashboard和报告至关重要，因为它能告知用户其分析的最新程度，并帮助管理对包含最新交易数据的预期。为何重要表明数据集的新鲜度，这对于理解流程 insights 和 dashboards 的及时性至关重要。获取方式这是在数据抽取、转换和加载（ETL）过程中生成并添加的元数据。示例 2024-05-21T02:00:00Z2024-05-22T02:00:00Z2024-05-23T02:00:00Z
	单据货币 DocumentCurrency	开具发票时使用的货币代码（例如：USD, EUR）。
描述此属性指定发票金额的货币类型。它存储在SAP的凭证抬头级别。凭证货币（Document Currency）对于任何财务分析都至关重要，以确保金额得到正确解读。它允许按货币筛选流程分析，并且是进行标准化报告和汇总时，将金额转换为通用本地货币的先决条件。为何重要提供解读财务金额所需的背景信息，并支持特定货币分析或转换为标准货币。获取方式源自 SAP 表 BKPF（会计凭证抬头）的 WAERS 字段。示例美元EURGBP
	发票处理时间 InvoiceProcessingTime	从首个活动（例如，发票创建）到最终活动（例如，付款）所经过的总耗时。
描述这是一个计算metric，用于衡量每张发票的端到端（end-to-end）周期时间。它通过取给定case（发票号）的最后一个事件的timestamp，减去第一个事件的timestamp而得出。此属性是“端到端发票周期时间”dashboard的主要KPI。它提供了衡量整体流程效率的高级指标。分析其平均值、中位数和分布有助于识别趋势，衡量流程改进的影响，并设定绩效目标。为何重要这是衡量整体流程效率的主要KPI。跟踪此KPI有助于识别改进措施对端到端流程生命周期的影响。获取方式计算字段：(案例中最后一个事件的时间戳) - (案例中第一个事件的时间戳)。示例 10天2小时15分钟5天0小时0分钟22天8小时30分钟
	收货单号 GoodsReceiptNumber	与基于采购订单的发票相关的收货凭证标识符。
描述此属性关联实物商品发票与收货 (GR) 单据。用于“采购订单/收货匹配绩效”dashboard，通过分析GR和发票过账时间，识别三方匹配流程中的延迟，优化实物交付与财务结算。为何重要完善了三方匹配（采购订单-收货-发票）的全貌，从而可以分析货物交付与发票处理之间的延迟。获取方式源自 SAP 表 RSEG（凭证项目，进项发票）的 LFBNR 字段（参考凭证编号）。示例 500000123450000015675000002100
	文档类型 DocumentType	SAP 中的一个代码，用于对会计凭证进行分类并控制其处理方式。
描述凭证类型用于区分各种业务交易，例如供应商发票（KR）、贷项凭证（KG）或MM发票（RE）。此分类管理着编号范围和可过账的账户类型等方面。在流程挖掘中，按凭证类型进行过滤可以实现更同质化的分析。例如，标准供应商发票（KR）的处理流程可能与贷项凭证（KG）的流程显著不同。分别分析这些凭证类型有助于防止误导性汇总，并为特定子流程提供更准确的洞察。为何重要支持按交易类型（例如，发票与贷项凭证）进行分析细分，从而获得更精准、更有针对性的流程洞察。获取方式源自 SAP 表 BKPF（会计凭证抬头）的 BLART 字段。示例 KRREKG
	是否已冲销 IsReversed	一个布尔标志，指示发票凭证是否已冲销。
描述此属性标记已被取消或冲销的发票，这通常是一个重要事件，表明存在严重错误或流程故障。在SAP系统中，已冲销凭证与冲销凭证是关联的。识别冲销发票对于评估流程质量和故障率至关重要。它有助于将这些末端异常情况从主流程中分离出来，从而进行专门的根本原因分析，以了解冲销发生的原因并制定预防措施。为何重要标记已取消的发票，这是流程失败的关键指标，有助于分析主要错误的根本原因。获取方式检查文档编号（BKPF-BELNR）是否存在于任何其他文档的BKPF-STBLG（冲销凭证）字段中。示例 truefalse
	是否已自动化 IsAutomated	一个标志，指示活动是由系统或批处理用户执行，还是由人工用户执行。
描述此布尔属性用于区分自动执行的活动（例如，通过batch job、EDI或OCR系统）与用户手动执行的活动。这通常通过分析与活动关联的username来确定。这对于“人工与自动化处理”dashboard至关重要。它有助于衡量自动化举措的成功程度，计算“人工干预率”KPI，并识别仍有待减少人工操作的机会。理解人与系统之间的接触点是提高效率的关键。为何重要有助于衡量流程的自动化水平，识别手动瓶颈，并追踪自动化举措的影响。获取方式通过检查用户ID（例如，CDHDR-USERNAME）是否属于预定义的系统、批处理或服务账户列表来推导。示例 truefalse
	是否逾期 IsOverdue	一个计算标志，如果发票在预定到期日之后支付，则为真。
描述此布尔属性通过比较“发票已清账（通过付款）”活动的日期与“发票到期日”而得出。如果付款日期晚于到期日，则此标志设为true。此属性直接支持“逾期付款率”KPI和“付款合规与折扣”dashboard。它方便了逾期付款的筛选和计数，有助于量化问题的规模。分析逾期发票的特征（例如，按供应商、公司代码）可以揭示根本原因。为何重要这项指标直接衡量付款条款的合规性，是逾期付款率 KPI 的重要依据，有助于改善供应商关系并避免罚款。获取方式计算字段：如果（付款日期 > 发票到期日）则为TRUE，否则为FALSE。示例 truefalse
	源系统 SourceSystem	识别提取数据的源 SAP ECC 系统。
描述此属性指定记录系统，例如SAP系统ID（SID），即发票处理数据源自何处。这在拥有多个ERP实例或混合系统的组织中尤为重要。在Process Mining分析中，它有助于区分来自不同系统的数据，这些系统可能具有独特的配置或流程变体。这使得系统间的比较分析成为可能，并确保data lineage（数据血缘）清晰，这对于data governance（数据治理）和验证至关重要。为何重要确保数据可追溯性，并支持在多 SAP 实例或其他源系统环境中进行流程分析。获取方式这通常是一个静态值，代表源ECC实例的SAP系统ID (SID)，在数据抽取过程中添加。示例 ECC_PROD_EUSAP_US_01E5P
	错失折扣 DiscountLost	一个计算标志，如果未利用可用的早期付款折扣，则为真。
描述此属性标识那些提供早期付款折扣，但实际付款却在折扣期结束后才进行的发票。其计算方式是，将付款日期与根据“付款条款”得出的折扣截止日期进行比较。这是衡量“早期付款折扣捕获率”的关键财务KPI。它揭示了错失的成本节约机会。通过分析折扣损失的原因（例如，漫长的审批周期、付款冻结），可以为具有明确投资回报的流程改进项目提供有力支持。为何重要揭示错失的财务机会，直接支持“提前付款折扣获取率”这一关键绩效指标 (KPI)，并为流程改进提供明确的经济驱动力。获取方式计算字段：如果（付款条件提供折扣）且（付款日期 > 折扣到期日）则为TRUE，否则为FALSE。示例 truefalse

必填推荐可选

应付账款发票处理活动

这些是您应在事件日志中记录的关键流程步骤和里程碑，以便准确发现和分析您的发票处理工作流。

6 推荐 7 可选

	活动	描述
	付款建议已创建	该发票已被纳入付款运行 (T-Code F110) 的付款建议清单。尽管这不是最终付款，但它表明了公司的付款意图。发票是根据其到期日和付款条款筛选出来的。
为何重要这是自动化付款流程中的第一步。从付款提案到最终执行之间的延迟，可能表明付款批次审批或调度存在问题。获取方式此事件通过在付款建议表（例如REGUP - 付款程序已处理项目）中找到发票凭证来识别。运行日期（REGUP-LAUFD）可用作事件时间戳。捕获在 REGUP 表中查找发票，并使用付款运行日期 (LAUFD)。事件类型 explicit
	发票凭证已创建	这标志着发票凭证在 SAP 系统中最初的创建（无论是暂存还是已正式过账）。这通常是发票的首个带时间戳的事件，也是流程分析的起点。该信息从 BKPF 表中的凭证录入日期和时间中获取。
为何重要此活动是发票处理流程的主要起始事件。分析从此刻起的时间有助于衡量总端到端周期时间，并识别初始数据录入延误。获取方式此事件是从会计凭证抬头记录的创建timestamp推断而来。具体而言，入账日期（BKPF-CPUDT）和入账时间（BKPF-CPUTM）的组合提供了该timestamp。捕获使用凭证抬头表BKPF中的创建timestamp (CPUDT, CPUTM)。事件类型 inferred
	发票已冲销	已过账的发票凭证通过创建冲销凭证而被取消。此操作抵消了原始发票的财务影响。这代表着流程的一个替代性（通常是负面的）终结。
为何重要冲销操作表明存在错误，例如数据录入不准确或发票重复。跟踪冲销的频率和原因，有助于发现提高数据质量和实现“一次性正确处理”的机会。获取方式此事件通过凭证抬头表（BKPF）识别。原始凭证将填充冲销凭证号（BKPF-STBLG）。事件时间是冲销凭证的过账日期。捕获在 BKPF.STBLG 中查找冲销凭证；从冲销凭证的过账日期获取时间戳。事件类型 explicit
	发票已审批	指定的审批人已确认发票有效并可用于支付。这可以是来自工作流系统的明确事件，也可以是从移除付款冻结中推断出来的。这是付款前的一个关键里程碑。
为何重要这是一个解除发票付款冻结的关键里程碑。在此步骤前后发生的延迟，可能分别指向审批人可用性不足或付款批次调度等问题。获取方式在 workflow 系统中，这是一个显式 event。否则，通常从移除付款冻结中推断出，并通过 BSEG-ZLSPR 字段的更改日志 (CDHDR/CDPOS) 捕获。捕获 workflow 系统或变更日志中关于解除付款冻结的事件。事件类型 explicit
	发票已过账	发票被正式记录在总账中，形成一项财务负债。暂存凭证会转换为已过账凭证，或者发票一开始就以已过账状态创建。这是财务会计中的一个基本步骤。
为何重要过账是一个关键里程碑，它使发票成为一项正式负债。它将数据录入阶段与流程中的主动付款管理阶段区分开来。获取方式从凭证抬头表 BKPF 中识别。已过账凭证的凭证状态 (BKPF-BSTAT) 为空，并且过账日期 (BKPF-BUDAT) 有效。事件 timestamp 为过账日期。捕获对于BKPF.BSTAT不为'V'（暂存）或'V'状态已被更改的凭证，请使用BKPF.BUDAT。事件类型 explicit
	发票已通过付款清算	发票已全额支付，未清项与付款凭证进行清账。这标志着该发票的应付账款流程成功完成。清账凭证信息记录了此过程。
为何重要这是主要成功路径的结束事件。达到此活动所需的时间是衡量整体流程效率的关键指标，对于计算付款及时性和折扣捕获也至关重要。获取方式当供应商行项目中填充了清账凭证号（BSEG-AUGBL）和清账日期（BSEG-AUGDT）时，即可识别此事件。清账日期即为该事件的timestamp。捕获当结算凭证(BSEG.AUGBL)有值时，请使用结算日期(BSEG.AUGDT)。事件类型 explicit
	付款冻结已解除	发票上之前设置的付款冻结已被移除，使其具备付款资格。这通常表示差异已解决或已获得批准。该事件从更改凭证表中捕获。
为何重要此活动表示异常已解决或审批已完成。冻结生效期间所耗费的时间代表着返工或等待时间，这是流程改进的关键领域。获取方式此事件通过追踪“付款冻结码”字段（BSEG-ZLSPR）的更改来捕获。表CDHDR和CDPOS中的更改日志记录了此字段何时被清除。捕获跟踪CDHDR/CDPOS中字段BSEG-ZLSPR更改为空白的更新条目。事件类型 explicit
	到期日已过但未付款	当当前日期超过发票的净到期日，但发票尚未清账时，就会发生此计算事件。它突显了有逾期风险或已经逾期的发票。这并非一个直接的系统事件。
为何重要此活动对于监控逾期付款率KPI至关重要。它主动标记需要立即关注的发票，以避免产生逾期付款费用并损害供应商关系。获取方式这是一个计算事件。它通过比较当前日期与所有未清（un-cleared）发票项的净到期日（BSEG-ZFBDT）而得出。该事件的timestamp即为到期日本身。捕获通过将当前日期与未清项目的BSEG.ZFBDT进行比较来推导。事件类型 calculated
	发票已暂存	指的是已输入SAP但尚未过账到总账的发票。这是一个明确的操作，旨在允许后续完成或审批。抬头表中的凭证状态字段会指示发票是否处于“暂存”状态。
为何重要暂存表示流程中的有意暂停，通常是为了进行审查或等待缺失信息。跟踪这一环节有助于在发票正式过账之前识别导致延迟的原因。获取方式从凭证抬头表 BKPF 中识别，其中凭证状态 (BKPF-BSTAT) 为 'V' (暂存)。事件时间为凭证创建时间 (BKPF-CPUDT, BKPF-CPUTM)。捕获筛选 BKPF.BSTAT = 'V' 的文档。事件类型 explicit
	发票已送审	发票已正式提交至审批workflow。此活动通常依赖于外部workflow系统或SAP中的特定状态变更。这通常标志着审批子流程的开始。
为何重要这标志着审批周期的开始。衡量从该事件到“发票已审批”的时间，对于“发票审批周期时间KPI”以及识别与审批人相关的瓶颈至关重要。获取方式如果使用 SAP Business Workflow，此 event 可以从 workflow 日志（例如 SWWLOG）中提取。否则，这通常是从因审批原因设置付款冻结而推断出的概念性步骤。捕获从 workflow 日志中提取，或根据特定付款冻结的设置进行推断。事件类型 explicit
	收货已匹配	表示通过将发票与收货凭证关联，来验证发票上的货物或服务是否已收到。这是三单匹配（Three-Way Match）的第三个环节。该状态通常从发票行项目中的引用信息推断得出。
为何重要与收货单进行匹配，旨在确认公司已收到其所支付款项对应的货物。跟踪这一环节有助于分析采购订单-收货-发票（PO-GR-Invoice）匹配时间的关键绩效指标（KPI），并简化验证流程。获取方式从 BSEG 表中的供应商行项目推断得出，通过检查已填充的收货单号 (BSEG-LFBNR) 或追溯关联采购订单的历史记录 (表 EKBE)。捕获检查BSEG.LFBNR中是否有收货（GR）引用，或通过EKBE中的采购订单历史记录进行检查。事件类型 inferred
	设置付款冻结	一张发票通常由于差异、待审批或其他问题，被主动冻结支付。这是发票行项目上设置的明确状态。该事件可以从变更凭证表中捕获。
为何重要此活动标志着异常或审批阶段的开始。分析付款冻结的频率和持续时间，是识别和解决流程瓶颈及争议的关键。获取方式该事件通过跟踪“付款冻结码”字段（BSEG-ZLSPR）的更改来捕获。表CDHDR和CDPOS中的更改日志会记录此字段何时被填充。捕获跟踪CDHDR/CDPOS中针对字段BSEG-ZLSPR的条目创建。事件类型 explicit
	采购订单已匹配	此活动表明发票行项目已成功关联到采购订单。这是三方匹配流程中的一个关键验证步骤。这通过发票行项目明细中是否存在采购订单号来推断。
为何重要采购订单匹配中的延迟可能是一个显著的瓶颈。分析此活动有助于衡量匹配流程的效率，并识别出偏离标准采购订单流程的发票。获取方式从 BSEG 表中的供应商行项目推断得出。如果采购订单号字段 (BSEG-EBELN) 已填充，则认为匹配在凭证创建时发生。捕获检查发票在BSEG.EBELN字段中是否有非空值。事件类型 inferred

推荐可选

提取指南

如何从 SAP ECC 获取您的数据

步骤

范围与设计：在编写代码之前，明确与您的应付账款流程相关的具体 SAP 公司代码（BUKRS）和凭证类型（BLART）。常见的 FI 发票凭证类型包括“KR”（供应商发票）和“RE”（总额发票）。这些将用于数据筛选。与业务利益相关者确认所有13项所需活动均与该流程相关。
创建ABAP程序：使用事务代码 SE38 或 SE80 创建一个新的可执行 ABAP 程序，例如 Z_PM_AP_EXTRACTION。提供一个描述性的标题。
定义选择屏幕：在程序中，定义一个选择屏幕，允许用户指定提取参数。这必须包括发票过账日期（P_BUDAT）、公司代码（P_BUKRS）和凭证类型（P_BLART）的日期范围。这使得工具能够灵活地进行不同分析。
声明数据结构：定义一个内部表结构，使其与最终事件日志格式匹配。该结构应包含 CaseID（发票号码）、Activity、EventTime 以及所有其他所需属性，如 VendorName、CompanyCode 等的字段。
选择基本发票数据：编写核心数据选择逻辑。程序将首先根据用户选择屏幕的参数，从 BKPF 中选择发票抬头，并从 BSEG 中选择行项目。高效的做法是在初始查询中将这些数据与供应商主数据表 LFA1 连接，以获取 VendorName。
生成“现状”事件：遍历所选的基本发票数据。对于每张发票，生成可以直接从 BKPF 和 BSEG 表中确定的事件。这包括“发票凭证已创建”、“发票已暂存”、“发票已过账”、“采购订单已匹配”、“收货已匹配”和“发票已通过付款清账”。将每个生成的事件作为新行添加到您的最终事件日志内部表。
生成更改日志事件：为了捕获付款冻结等更改，从更改凭证表 CDHDR（抬头）和 CDPOS（行项目）中选择数据。在 BSEG 表中搜索 ZLSPR（付款冻结码）字段的更改。遍历这些更改凭证以创建“付款冻结已设置”和“付款冻结已移除”事件，并带有 CDHDR 中的精确时间戳。
生成付款运行事件：查询付款建议表 REGUH（抬头）和 REGUP（行项目），以识别发票何时被纳入付款建议中。这将生成“付款建议已创建”活动。
生成计算和推断事件：针对非直接系统录入的事件实现逻辑。对于“付款截止日期已过未付款”，比较发票到期日（BSEG-NETDT）与清账日期或当前日期。对于“发票已批准”，您可能需要从移除特定的“需要审批”付款冻结中推断出来。
整合和格式化数据：确保所有时间戳都转换为一致的格式（例如，YYYY-MM-DD HH:MM:SS）。将上述步骤生成的所有事件合并到最终的单个内部表中。
创建输出文件：使用 cl_gui_frontend_services=>gui_download 方法或将文件写入应用服务器。输出应为扁平文件（例如，CSV），其标题行与 ProcessMind 数据模型要求匹配。该文件现在已准备好上传。

配置

选择参数：程序应具有一个选择屏幕，其中包含以下必填和可选字段：
- S_BUDAT：过账日期范围（必填）。定义数据提取的时间窗口。
- S_BUKRS：公司代码（必填）。将数据提取集中于特定的法律实体。
- S_BLART：凭证类型（可选）。允许筛选特定发票类型，如KR或RE。
日期范围建议：对于初始数据提取，建议选择最近3到6个月的期间，以确保数据量可控并加快执行速度。完整的历史数据加载应在后台运行。
性能考量：从更改日志表（CDHDR/CDPOS）中进行选择可能会很慢。请确保查询尽可能使用主键和索引字段。对于非常大的数据集，请考虑将程序作为后台作业（事务代码SM36）运行，以避免对话进程超时。
前提条件与权限：
- 执行程序的用户需要访问表BKPF、BSEG、LFA1、CDHDR、CDPOS、REGUH和REGUP的权限。
- 运行程序需要事务代码SE38的执行权限。
- 若要捕获“付款冻结设置/解除”活动，BSEG表ZLSPR字段的更改凭证日志记录必须处于启用状态。这是一项系统配置设置。

a 查询示例 abap

`abap
*&---------------------------------------------------------------------*
*& Report Z_PM_AP_EXTRACTION
*&---------------------------------------------------------------------*
*& This program extracts Accounts Payable invoice data for process mining.
*& It creates an event log based on key business activities.
*&---------------------------------------------------------------------*
REPORT z_pm_ap_extraction.

*&---------------------------------------------------------------------*
*& Global Data Declarations
*&---------------------------------------------------------------------*
TABLES: bkpf, bseg, lfa1.

TYPES:
  BEGIN OF ty_event_log,
    invoicenumber       TYPE belnr_d,
    activity            TYPE string,
    eventtime           TYPE string,
    processoruser       TYPE uname,
    vendorname          TYPE name1_gp,
    companycode         TYPE bukrs,
    invoicegrossamount  TYPE wrbtr,
    invoiceduedate      TYPE c, "char20
    paymentterms        TYPE char256,
    purchaseordernumber TYPE ebeln,
    paymentblockreason  TYPE char256,
    approver            TYPE uname,
  END OF ty_event_log.

DATA: gt_event_log TYPE TABLE OF ty_event_log.

TYPES:
  BEGIN OF ty_invoice_data,
    bukrs TYPE bkpf-bukrs,
    belnr TYPE bkpf-belnr,
    gjahr TYPE bkpf-gjahr,
    blart TYPE bkpf-blart,
    budat TYPE bkpf-budat,
    cpudt TYPE bkpf-cpudt,
    cputm TYPE bkpf-cputm,
    usnam TYPE bkpf-usnam,
    bstat TYPE bkpf-bstat,
    stblg TYPE bkpf-stblg,
    xblnr TYPE bkpf-xblnr,
    lifnr TYPE bseg-lifnr,
    buzei TYPE bseg-buzei,
    wrbtr TYPE bseg-wrbtr,
    zterm TYPE bseg-zterm,
    zlspr TYPE bseg-zlspr,
    netdt TYPE bseg-netdt,
    ebeln TYPE bseg-ebeln,
    lfbnr TYPE bseg-lfbnr,
    augbl TYPE bseg-augbl,
    augdt TYPE bseg-augdt,
    name1 TYPE lfa1-name1,
  END OF ty_invoice_data.

DATA: gt_invoice_data TYPE TABLE OF ty_invoice_data.

*&---------------------------------------------------------------------*
*& Selection Screen
*&---------------------------------------------------------------------*
SELECTION-SCREEN BEGIN OF BLOCK b1 WITH FRAME TITLE TEXT-001.
  SELECT-OPTIONS: s_budat FOR bkpf-budat OBLIGATORY,  " Posting Date
                    s_bukrs FOR bkpf-bukrs OBLIGATORY,  " Company Code
                    s_blart FOR bkpf-blart.             " Document Type
SELECTION-SCREEN END OF BLOCK b1.

PARAMETERS: p_file TYPE rlgrap-filename OBLIGATORY DEFAULT 'C:\temp\ap_event_log.csv'.

*&---------------------------------------------------------------------*
*& Main Program Logic
*&---------------------------------------------------------------------*
START-OF-SELECTION.

  " 1. Select base invoice data
  SELECT
    h~bukrs, h~belnr, h~gjahr, h~blart, h~budat, h~cpudt, h~cputm, h~usnam,
    h~bstat, h~stblg, h~xblnr, i~lifnr, i~buzei, i~wrbtr, i~zterm, i~zlspr,
    i~netdt, i~ebeln, i~lfbnr, i~augbl, i~augdt, v~name1
    INTO TABLE @gt_invoice_data
    FROM bkpf AS h
    INNER JOIN bseg AS i ON  h~bukrs = i~bukrs
                         AND h~belnr = i~belnr
                         AND h~gjahr = i~gjahr
    INNER JOIN lfa1 AS v ON i~lifnr = v~lifnr
    WHERE h~budat IN @s_budat
      AND h~bukrs IN @s_bukrs
      AND h~blart IN @s_blart
      AND i~koart = 'K'. " Vendor line items only

  CHECK gt_invoice_data IS NOT INITIAL.

  " 2. Process each invoice to generate events
  LOOP AT gt_invoice_data ASSIGNING FIELD-SYMBOL(<fs_invoice>).
    " Generate events for each invoice document/line item
    PERFORM generate_events USING <fs_invoice>.
  ENDLOOP.

  " 3. Generate events from change documents (payment blocks)
  PERFORM generate_change_log_events.

  " 4. Generate events from payment proposal run
  PERFORM generate_payment_proposal_events.

  " 5. Download the final event log
  PERFORM download_event_log.


*&---------------------------------------------------------------------*
*&      Form  GENERATE_EVENTS
*&---------------------------------------------------------------------*
FORM generate_events USING is_invoice TYPE ty_invoice_data.
  DATA: ls_event_log TYPE ty_event_log,
        lv_timestamp TYPE string.

  " Common attributes for all events of this invoice
  ls_event_log-invoicenumber      = is_invoice-belnr.
  ls_event_log-companycode        = is_invoice-bukrs.
  ls_event_log-vendorname         = is_invoice-name1.
  ls_event_log-invoicegrossamount = is_invoice-wrbtr.
  ls_event_log-paymentterms       = is_invoice-zterm.
  ls_event_log-purchaseordernumber= is_invoice-ebeln.

  " Format Due Date
  WRITE is_invoice-netdt TO ls_event_log-invoiceduedate DD/MM/YYYY.

  " Activity: Invoice Document Created
  CONCATENATE is_invoice-cpudt is_invoice-cputm INTO lv_timestamp.
  ls_event_log-activity     = 'Invoice Document Created'.
  ls_event_log-eventtime    = lv_timestamp.
  ls_event_log-processoruser= is_invoice-usnam.
  APPEND ls_event_log TO gt_event_log.

  " Activity: Invoice Parked
  IF is_invoice-bstat = 'V'.
    ls_event_log-activity     = 'Invoice Parked'.
    ls_event_log-eventtime    = lv_timestamp. " Parked at time of creation
    ls_event_log-processoruser= is_invoice-usnam.
    APPEND ls_event_log TO gt_event_log.
  ENDIF.

  " Activity: Invoice Posted
  IF is_invoice-bstat = space.
    CONCATENATE is_invoice-budat '000000' INTO lv_timestamp.
    ls_event_log-activity     = 'Invoice Posted'.
    ls_event_log-eventtime    = lv_timestamp.
    ls_event_log-processoruser= is_invoice-usnam.
    APPEND ls_event_log TO gt_event_log.
  ENDIF.

  " Activity: Purchase Order Matched
  IF is_invoice-ebeln IS NOT INITIAL.
    CONCATENATE is_invoice-cpudt is_invoice-cputm INTO lv_timestamp.
    ls_event_log-activity     = 'Purchase Order Matched'.
    ls_event_log-eventtime    = lv_timestamp.
    ls_event_log-processoruser= is_invoice-usnam.
    APPEND ls_event_log TO gt_event_log.
  ENDIF.

  " Activity: Goods Receipt Matched
  IF is_invoice-lfbnr IS NOT INITIAL.
    CONCATENATE is_invoice-cpudt is_invoice-cputm INTO lv_timestamp.
    ls_event_log-activity     = 'Goods Receipt Matched'.
    ls_event_log-eventtime    = lv_timestamp.
    ls_event_log-processoruser= is_invoice-usnam.
    APPEND ls_event_log TO gt_event_log.
  ENDIF.

  " Activity: Invoice Cleared By Payment
  IF is_invoice-augbl IS NOT INITIAL.
    CONCATENATE is_invoice-augdt '000000' INTO lv_timestamp.
    ls_event_log-activity     = 'Invoice Cleared By Payment'.
    ls_event_log-eventtime    = lv_timestamp.
    " Processor user for clearing is not in this structure, would require another SELECT
    ls_event_log-processoruser = ''.
    APPEND ls_event_log TO gt_event_log.
  ENDIF.

  " Activity: Invoice Reversed
  IF is_invoice-stblg IS NOT INITIAL.
    SELECT SINGLE budat INTO @DATA(lv_rev_date) FROM bkpf
      WHERE belnr = @is_invoice-stblg AND gjahr = @is_invoice-gjahr.
    IF sy-subrc = 0.
      CONCATENATE lv_rev_date '000000' INTO lv_timestamp.
      ls_event_log-activity     = 'Invoice Reversed'.
      ls_event_log-eventtime    = lv_timestamp.
      ls_event_log-processoruser= sy-uname.
      APPEND ls_event_log TO gt_event_log.
    ENDIF.
  ENDIF.

  " Activity: Due Date Passed Without Payment
  IF is_invoice-netdt < sy-datum AND is_invoice-augbl IS INITIAL.
    CONCATENATE is_invoice-netdt '000000' INTO lv_timestamp.
    ls_event_log-activity      = 'Due Date Passed Without Payment'.
    ls_event_log-eventtime     = lv_timestamp.
    ls_event_log-processoruser = 'SYSTEM'.
    APPEND ls_event_log TO gt_event_log.
  ENDIF.

ENDFORM.

*&---------------------------------------------------------------------*
*&      Form  GENERATE_CHANGE_LOG_EVENTS
*&---------------------------------------------------------------------*
FORM generate_change_log_events.
  TYPES: BEGIN OF ty_cdpos,
           objectid TYPE cdpos-objectid,
           changenr TYPE cdpos-changenr,
           fname    TYPE cdpos-fname,
           value_new TYPE cdpos-value_new,
           value_old TYPE cdpos-value_old,
         END OF ty_cdpos.

  DATA: lt_cdpos TYPE TABLE OF ty_cdpos,
        lt_cdhdr TYPE TABLE OF cdhdr.

  DATA: ls_event_log TYPE ty_event_log.

  DATA(lt_invoice_range) = VALUE range_belnr_d( FOR wa IN gt_invoice_data ( sign = 'I' option = 'EQ' low = wa-belnr ) ).
  CHECK lt_invoice_range IS NOT INITIAL.

  SELECT objectclas, objectid, changenr, username, udate, utime
    FROM cdhdr
    INTO TABLE @lt_cdhdr
    FOR ALL ENTRIES IN @gt_invoice_data
    WHERE objectclas = 'BELEG'
      AND objectid   = @gt_invoice_data-objectid.
  CHECK sy-subrc = 0.

  SELECT objectid, changenr, fname, value_new, value_old
    FROM cdpos
    INTO TABLE @lt_cdpos
    FOR ALL ENTRIES IN @lt_cdhdr
    WHERE changenr = @lt_cdhdr-changenr
      AND objectid = @lt_cdhdr-objectid
      AND tabname  = 'BSEG'
      AND fname    = 'ZLSPR'.
  CHECK sy-subrc = 0.

  LOOP AT lt_cdpos ASSIGNING FIELD-SYMBOL(<fs_cdpos>).
    READ TABLE lt_cdhdr ASSIGNING FIELD-SYMBOL(<fs_cdhdr>) WITH KEY changenr = <fs_cdpos>-changenr.
    IF sy-subrc <> 0. CONTINUE. ENDIF.

    " Find corresponding invoice data
    DATA(lv_belnr) = <fs_cdhdr>-objectid+10(10).
    READ TABLE gt_invoice_data ASSIGNING FIELD-SYMBOL(<fs_invoice>) WITH KEY belnr = lv_belnr.
    IF sy-subrc <> 0. CONTINUE. ENDIF.

    " Populate common event data
    ls_event_log-invoicenumber       = <fs_invoice>-belnr.
    ls_event_log-companycode         = <fs_invoice>-bukrs.
    ls_event_log-vendorname          = <fs_invoice>-name1.
    ls_event_log-invoicegrossamount  = <fs_invoice>-wrbtr.
    ls_event_log-purchaseordernumber = <fs_invoice>-ebeln.
    ls_event_log-processoruser       = <fs_cdhdr>-username.

    CONCATENATE <fs_cdhdr>-udate <fs_cdhdr>-utime INTO ls_event_log-eventtime.

    " Activity: Payment Block Set
    IF <fs_cdpos>-value_new IS NOT INITIAL.
      ls_event_log-activity           = 'Payment Block Set'.
      ls_event_log-paymentblockreason = <fs_cdpos>-value_new.
      APPEND ls_event_log TO gt_event_log.
    ENDIF.

    " Activity: Payment Block Removed
    IF <fs_cdpos>-value_new IS INITIAL AND <fs_cdpos>-value_old IS NOT INITIAL.
      ls_event_log-activity           = 'Payment Block Removed'.
      ls_event_log-paymentblockreason = <fs_cdpos>-value_old.
      APPEND ls_event_log TO gt_event_log.

      " INFERRED Activity: Invoice Approved
      " Assumption: Removing a specific block (e.g., 'A' for Approval) means approval.
      IF <fs_cdpos>-value_old = '[Your Approval Block Reason]'. " E.g., 'A'
        ls_event_log-activity           = 'Invoice Approved'.
        ls_event_log-approver           = <fs_cdhdr>-username.
        APPEND ls_event_log TO gt_event_log.
      ENDIF.
    ENDIF.

    " INFERRED Activity: Invoice Routed For Approval
    " Assumption: Setting a specific block (e.g., 'A' for Approval) means routed for approval.
    IF <fs_cdpos>-value_new = '[Your Approval Block Reason]'. " E.g., 'A'
      ls_event_log-activity           = 'Invoice Routed For Approval'.
      ls_event_log-approver           = ''.
      APPEND ls_event_log TO gt_event_log.
    ENDIF.

  ENDLOOP.
ENDFORM.

*&---------------------------------------------------------------------*
*&      Form  GENERATE_PAYMENT_PROPOSAL_EVENTS
*&---------------------------------------------------------------------*
FORM generate_payment_proposal_events.
  DATA: lt_regup TYPE TABLE OF regup,
        ls_reguh TYPE reguh.
  DATA: ls_event_log TYPE ty_event_log.

  SELECT * FROM regup INTO TABLE @lt_regup
    FOR ALL ENTRIES IN @gt_invoice_data
    WHERE bukrs = @gt_invoice_data-bukrs
      AND belnr = @gt_invoice_data-belnr
      AND gjahr = @gt_invoice_data-gjahr.
  CHECK sy-subrc = 0.

  LOOP AT lt_regup ASSIGNING FIELD-SYMBOL(<fs_regup>).
    READ TABLE gt_invoice_data ASSIGNING FIELD-SYMBOL(<fs_invoice>) WITH KEY belnr = <fs_regup>-belnr.
    IF sy-subrc <> 0. CONTINUE. ENDIF.

    SELECT SINGLE * FROM reguh INTO @ls_reguh
      WHERE rundat = @<fs_regup>-rundat AND runid = @<fs_regup>-runid.
    IF sy-subrc <> 0. CONTINUE. ENDIF.

    ls_event_log-invoicenumber      = <fs_invoice>-belnr.
    ls_event_log-companycode        = <fs_invoice>-bukrs.
    ls_event_log-vendorname         = <fs_invoice>-name1.
    ls_event_log-invoicegrossamount = <fs_invoice>-wrbtr.
    ls_event_log-processoruser      = ls_reguh-usnam.
    ls_event_log-activity           = 'Payment Proposal Created'.

    CONCATENATE ls_reguh-rundat '000000' INTO ls_event_log-eventtime.
    APPEND ls_event_log TO gt_event_log.
  ENDLOOP.
ENDFORM.

*&---------------------------------------------------------------------*
*&      Form  DOWNLOAD_EVENT_LOG
*&---------------------------------------------------------------------*
FORM download_event_log.
  DATA: lt_final_output TYPE TABLE OF string,
        l_header        TYPE string.

  SORT gt_event_log BY invoicenumber eventtime.

  " Create Header
  l_header = "InvoiceNumber","Activity","EventTime","ProcessorUser","VendorName","CompanyCode","InvoiceGrossAmount","InvoiceDueDate","PaymentTerms","PurchaseOrderNumber","PaymentBlockReason","Approver".
  APPEND l_header TO lt_final_output.

  " Create Data Rows
  LOOP AT gt_event_log ASSIGNING FIELD-SYMBOL(<fs_log>).
    DATA(l_line) = |"{ <fs_log>-invoicenumber }","{
                     <fs_log>-activity }","{
                     <fs_log>-eventtime }","{
                     <fs_log>-processoruser }","{
                     <fs_log>-vendorname }","{
                     <fs_log>-companycode }","{
                     <fs_log>-invoicegrossamount }","{
                     <fs_log>-invoiceduedate }","{
                     <fs_log>-paymentterms }","{
                     <fs_log>-purchaseordernumber }","{
                     <fs_log>-paymentblockreason }","{
                     <fs_log>-approver }"|.
    APPEND l_line TO lt_final_output.
  ENDLOOP.

  " Download file to presentation server
  cl_gui_frontend_services=>gui_download(
    EXPORTING
      filename = p_file
      filetype = 'ASC'
    CHANGING
      data_tab = lt_final_output
    EXCEPTIONS
      OTHERS   = 4 ).

  IF sy-subrc <> 0.
    MESSAGE 'File download failed.' TYPE 'E'.
  ELSE.
    MESSAGE |File '{ p_file }' downloaded successfully.| TYPE 'I'.
  ENDIF.
ENDFORM.
`

步骤

前提条件和访问权限：确保您的数据库用户拥有读取所需 SAP 表（例如 BKPF、BSEG、LFA1、CDHDR、CDPOS、REGUH、REGUP）的权限。此方法需要直接访问数据库，可能需要您的 IT 安全团队和 SAP Basis 团队的批准。
确认系统详情：确认您的 SAP ECC 系统使用的具体数据库类型（例如 SAP HANA、Oracle、MS SQL Server）。提供的 SQL 查询采用 SAP HANA 上常见的语法，对于其他数据库，特别是时间戳的创建，可能需要进行微调。
定制查询：在数据库客户端（如 DBeaver、SAP HANA Studio 或 SQL Server Management Studio）中打开提供的 SQL 查询。修改查询中 BaseInvoices CTE 顶部的占位符值：
- '{StartDate}'：设置提取周期的开始日期（例如，'2023-01-01'）。
- '{EndDate}'：设置提取周期的结束日期（例如，'2023-06-30'）。
- '{CompanyCodes}'：替换为您要分析的公司代码的逗号分隔列表（例如，'1000'、'2000'）。
- '{DocTypes}'：确认系统中供应商发票的凭证类型。'KR' 和 'RE' 是常见的默认值。
执行查询：针对 SAP 数据库运行定制的 SQL 查询。执行时间将取决于日期范围、公司代码数量和整体系统规模。
审查查询输出：查询完成后，检查输出的前几百行。确认所有列（InvoiceNumber、Activity、EventTime 等）都存在并按预期填充。
处理潜在错误：如果遇到语法错误，请调整日期/时间函数以匹配您的特定数据库。例如，TO_TIMESTAMP (HANA/Oracle) 可能需要更改为 CAST 或 CONVERT 函数 (SQL Server)。
导出数据：将 SQL 客户端中的完整结果集导出为 CSV（逗号分隔值）文件。确保文件编码设置为 UTF-8，以防止特殊字符问题。
准备上传：将 CSV 文件中的列名与查询别名（InvoiceNumber、Activity、EventTime、ProcessorUser 等）精确对应。这将确保在上传过程中实现无缝映射。
上传至流程挖掘工具：将最终的 CSV 文件上传到您的流程挖掘应用程序。将 InvoiceNumber 列映射到“案例ID”，将 Activity 映射到“活动”，将 EventTime 映射到“时间戳”，即可开始您的分析。

配置

数据库连接：需要与底层SAP ECC数据库建立直接的只读连接。您需要准备服务器地址、数据库名称、端口、用户名和密码。
日期范围：建议初期设定有限的日期范围（例如3-6个月），以确保性能并验证结果，之后再执行大规模数据提取。日期筛选将应用于凭证创建日期（BKPF.CPUDT）。
公司代码：查询必须通过一个或多个公司代码（BUKRS）值进行筛选。不建议在大型系统上一次性提取所有公司代码的数据。
凭证类型：数据提取将按发票凭证类型（BLART）进行筛选。供应商发票的常见类型为KR（供应商发票）和RE（发票收据）。请咨询您的财务团队，以确认适用于您组织的凭证类型。
表权限：数据库用户需要对以下表具有SELECT查询权限：BKPF、BSEG、LFA1、CDHDR、CDPOS、REGUH、REGUP。
时间戳转换：查询使用TO_TIMESTAMP(CONCAT(date, time), 'YYYYMMDDHH24MISS')。这与SAP HANA和Oracle兼容。对于SQL Server，您可能需要使用CAST(CONCAT(date, ' ', time) AS DATETIME2)。具体实现可能需要根据您的目标数据库进行调整。

a 查询示例 sql

`sql
WITH BaseInvoices AS (
  SELECT
    BKPF.MANDT,
    BKPF.BUKRS,
    BKPF.BELNR,
    BKPF.GJAHR,
    CONCAT(CONCAT(CONCAT(BKPF.BUKRS, '-'), BKPF.BELNR), BKPF.GJAHR) AS InvoiceNumber,
    BKPF.CPUDT AS CreationDate,
    BKPF.CPUTM AS CreationTime,
    BKPF.USNAM AS CreatorUser,
    BKPF.BSTAT AS DocStatus,
    BKPF.STBLG AS ReversalDoc,
    BKPF.STGRD AS ReversalDate
  FROM BKPF
  WHERE
    BKPF.BUKRS IN ('{CompanyCodes}') -- E.g., '1000', '2000'
    AND BKPF.BLART IN ('{DocTypes}') -- E.g., 'KR', 'RE'
    AND BKPF.CPUDT BETWEEN '{StartDate}' AND '{EndDate}' -- E.g., '20230101' AND '20231231'
),
InvoiceVendorLines AS (
  SELECT
    BSEG.MANDT, BSEG.BUKRS, BSEG.BELNR, BSEG.GJAHR,
    MAX(BSEG.LIFNR) AS VendorNumber,
    MAX(BSEG.ZFBDT) AS DueDate,
    MAX(BSEG.ZTERM) AS PaymentTerms,
    MAX(BSEG.ZLSPR) AS PaymentBlockReason,
    SUM(CASE WHEN BSEG.SHKZG = 'S' THEN BSEG.WRBTR ELSE -BSEG.WRBTR END) AS InvoiceGrossAmount,
    MAX(BSEG.EBELN) AS PurchaseOrderNumber,
    MAX(BSEG.AUGDT) AS ClearingDate,
    MAX(AUGBL) AS ClearingDoc
  FROM BSEG
  INNER JOIN BaseInvoices ON BSEG.MANDT = BaseInvoices.MANDT
    AND BSEG.BUKRS = BaseInvoices.BUKRS
    AND BSEG.BELNR = BaseInvoices.BELNR
    AND BSEG.GJAHR = BaseInvoices.GJAHR
  WHERE BSEG.KOART = 'K' -- Filter for Vendor line items
  GROUP BY BSEG.MANDT, BSEG.BUKRS, BSEG.BELNR, BSEG.GJAHR
)
-- 1. Invoice Document Created
SELECT
  bi.InvoiceNumber AS "InvoiceNumber",
  'Invoice Document Created' AS "Activity",
  TO_TIMESTAMP(CONCAT(bi.CreationDate, bi.CreationTime), 'YYYYMMDDHH24MISS') AS "EventTime",
  bi.CreatorUser AS "ProcessorUser",
  vend.Name1 AS "VendorName",
  bi.BUKRS AS "CompanyCode",
  ivl.InvoiceGrossAmount AS "InvoiceGrossAmount",
  ivl.DueDate AS "InvoiceDueDate",
  ivl.PaymentTerms AS "PaymentTerms",
  ivl.PurchaseOrderNumber AS "PurchaseOrderNumber",
  ivl.PaymentBlockReason AS "PaymentBlockReason",
  NULL AS "Approver"
FROM BaseInvoices bi
JOIN InvoiceVendorLines ivl ON bi.MANDT = ivl.MANDT AND bi.BUKRS = ivl.BUKRS AND bi.BELNR = ivl.BELNR AND bi.GJAHR = ivl.GJAHR
LEFT JOIN LFA1 vend ON ivl.MANDT = vend.MANDT AND ivl.VendorNumber = vend.LIFNR

UNION ALL

-- 2. Invoice Parked
SELECT
  bi.InvoiceNumber,
  'Invoice Parked' AS "Activity",
  TO_TIMESTAMP(CONCAT(bi.CreationDate, bi.CreationTime), 'YYYYMMDDHH24MISS') AS "EventTime",
  bi.CreatorUser AS "ProcessorUser",
  vend.Name1 AS "VendorName",
  bi.BUKRS AS "CompanyCode",
  ivl.InvoiceGrossAmount AS "InvoiceGrossAmount",
  ivl.DueDate AS "InvoiceDueDate",
  ivl.PaymentTerms AS "PaymentTerms",
  ivl.PurchaseOrderNumber AS "PurchaseOrderNumber",
  ivl.PaymentBlockReason AS "PaymentBlockReason",
  NULL AS "Approver"
FROM BaseInvoices bi
JOIN InvoiceVendorLines ivl ON bi.MANDT = ivl.MANDT AND bi.BUKRS = ivl.BUKRS AND bi.BELNR = ivl.BELNR AND bi.GJAHR = ivl.GJAHR
LEFT JOIN LFA1 vend ON ivl.MANDT = vend.MANDT AND ivl.VendorNumber = vend.LIFNR
WHERE bi.DocStatus = 'V'

UNION ALL

-- 3. Purchase Order Matched & 4. Goods Receipt Matched (Inferred at creation)
SELECT
  bi.InvoiceNumber,
  CASE WHEN ivl.PurchaseOrderNumber IS NOT NULL THEN 'Purchase Order Matched' ELSE 'Goods Receipt Matched' END AS "Activity",
  TO_TIMESTAMP(CONCAT(bi.CreationDate, bi.CreationTime), 'YYYYMMDDHH24MISS') AS "EventTime",
  bi.CreatorUser AS "ProcessorUser",
  vend.Name1 AS "VendorName",
  bi.BUKRS AS "CompanyCode",
  ivl.InvoiceGrossAmount AS "InvoiceGrossAmount",
  ivl.DueDate AS "InvoiceDueDate",
  ivl.PaymentTerms AS "PaymentTerms",
  ivl.PurchaseOrderNumber AS "PurchaseOrderNumber",
  ivl.PaymentBlockReason AS "PaymentBlockReason",
  NULL AS "Approver"
FROM BaseInvoices bi
JOIN InvoiceVendorLines ivl ON bi.MANDT = ivl.MANDT AND bi.BUKRS = ivl.BUKRS AND bi.BELNR = ivl.BELNR AND bi.GJAHR = ivl.GJAHR
LEFT JOIN LFA1 vend ON ivl.MANDT = vend.MANDT AND ivl.VendorNumber = vend.LIFNR
WHERE ivl.PurchaseOrderNumber IS NOT NULL -- Simplified logic: If PO exists, we assume match for both.

UNION ALL

-- 5. Payment Block Set & 6. Invoice Routed For Approval (Proxy) & 8. Payment Block Removed
SELECT
  bi.InvoiceNumber,
  CASE 
    WHEN pos.VALUE_OLD = ' ' AND pos.VALUE_NEW != ' ' AND pos.VALUE_NEW = 'A' THEN 'Invoice Routed For Approval'
    WHEN pos.VALUE_OLD = ' ' AND pos.VALUE_NEW != ' ' THEN 'Payment Block Set'
    ELSE 'Payment Block Removed' 
  END AS "Activity",
  TO_TIMESTAMP(CONCAT(hdr.UDATE, hdr.UTIME), 'YYYYMMDDHH24MISS') AS "EventTime",
  hdr.USERNAME AS "ProcessorUser",
  vend.Name1 AS "VendorName",
  bi.BUKRS AS "CompanyCode",
  ivl.InvoiceGrossAmount AS "InvoiceGrossAmount",
  ivl.DueDate AS "InvoiceDueDate",
  ivl.PaymentTerms AS "PaymentTerms",
  ivl.PurchaseOrderNumber AS "PurchaseOrderNumber",
  pos.VALUE_NEW AS "PaymentBlockReason",
  NULL AS "Approver"
FROM CDHDR hdr
JOIN CDPOS pos ON hdr.MANDANT = pos.MANDANT AND hdr.OBJECTCLAS = pos.OBJECTCLAS AND hdr.OBJECTID = pos.OBJECTID AND hdr.CHANGENR = pos.CHANGENR
JOIN BaseInvoices bi ON hdr.MANDANT = bi.MANDT AND hdr.OBJECTID = CONCAT(bi.BUKRS, bi.BELNR, bi.GJAHR)
JOIN InvoiceVendorLines ivl ON bi.MANDT = ivl.MANDT AND bi.BUKRS = ivl.BUKRS AND bi.BELNR = ivl.BELNR AND bi.GJAHR = ivl.GJAHR
LEFT JOIN LFA1 vend ON ivl.MANDT = vend.MANDT AND ivl.VendorNumber = vend.LIFNR
WHERE hdr.OBJECTCLAS = 'BELEG' AND pos.TABNAME = 'BSEG' AND pos.FNAME = 'ZLSPR' AND pos.VALUE_NEW != pos.VALUE_OLD

UNION ALL

-- 7. Invoice Approved (Proxy by removal of 'A' block)
SELECT
  bi.InvoiceNumber,
  'Invoice Approved' AS "Activity",
  TO_TIMESTAMP(CONCAT(hdr.UDATE, hdr.UTIME), 'YYYYMMDDHH24MISS') AS "EventTime",
  hdr.USERNAME AS "ProcessorUser", -- This user is the approver
  vend.Name1 AS "VendorName",
  bi.BUKRS AS "CompanyCode",
  ivl.InvoiceGrossAmount AS "InvoiceGrossAmount",
  ivl.DueDate AS "InvoiceDueDate",
  ivl.PaymentTerms AS "PaymentTerms",
  ivl.PurchaseOrderNumber AS "PurchaseOrderNumber",
  pos.VALUE_NEW AS "PaymentBlockReason",
  hdr.USERNAME AS "Approver"
FROM CDHDR hdr
JOIN CDPOS pos ON hdr.MANDANT = pos.MANDANT AND hdr.OBJECTCLAS = pos.OBJECTCLAS AND hdr.OBJECTID = pos.OBJECTID AND hdr.CHANGENR = pos.CHANGENR
JOIN BaseInvoices bi ON hdr.MANDANT = bi.MANDT AND hdr.OBJECTID = CONCAT(bi.BUKRS, bi.BELNR, bi.GJAHR)
JOIN InvoiceVendorLines ivl ON bi.MANDT = ivl.MANDT AND bi.BUKRS = ivl.BUKRS AND bi.BELNR = ivl.BELNR AND bi.GJAHR = ivl.GJAHR
LEFT JOIN LFA1 vend ON ivl.MANDT = vend.MANDT AND ivl.VendorNumber = vend.LIFNR
WHERE hdr.OBJECTCLAS = 'BELEG' AND pos.TABNAME = 'BSEG' AND pos.FNAME = 'ZLSPR' AND pos.VALUE_OLD = 'A' AND pos.VALUE_NEW = ' '

UNION ALL

-- 9. Invoice Posted
SELECT
  bi.InvoiceNumber,
  'Invoice Posted' AS "Activity",
  CASE -- If posted from parked, use change date. If posted directly, use creation date.
    WHEN chg.EventTime IS NOT NULL THEN chg.EventTime
    ELSE TO_TIMESTAMP(CONCAT(bi.CreationDate, bi.CreationTime), 'YYYYMMDDHH24MISS')
  END AS "EventTime",
  CASE
    WHEN chg.USERNAME IS NOT NULL THEN chg.USERNAME
    ELSE bi.CreatorUser
  END AS "ProcessorUser",
  vend.Name1 AS "VendorName",
  bi.BUKRS AS "CompanyCode",
  ivl.InvoiceGrossAmount AS "InvoiceGrossAmount",
  ivl.DueDate AS "InvoiceDueDate",
  ivl.PaymentTerms AS "PaymentTerms",
  ivl.PurchaseOrderNumber AS "PurchaseOrderNumber",
  ivl.PaymentBlockReason AS "PaymentBlockReason",
  NULL AS "Approver"
FROM BaseInvoices bi
JOIN InvoiceVendorLines ivl ON bi.MANDT = ivl.MANDT AND bi.BUKRS = ivl.BUKRS AND bi.BELNR = ivl.BELNR AND bi.GJAHR = ivl.GJAHR
LEFT JOIN LFA1 vend ON ivl.MANDT = vend.MANDT AND ivl.VendorNumber = vend.LIFNR
LEFT JOIN (
    SELECT hdr.MANDANT, hdr.OBJECTID, hdr.USERNAME, TO_TIMESTAMP(CONCAT(hdr.UDATE, hdr.UTIME), 'YYYYMMDDHH24MISS') AS EventTime
    FROM CDHDR hdr JOIN CDPOS pos ON hdr.CHANGENR = pos.CHANGENR AND hdr.OBJECTID = pos.OBJECTID
    WHERE hdr.OBJECTCLAS = 'BELEG' AND pos.TABNAME = 'BKPF' AND pos.FNAME = 'BSTAT' AND pos.VALUE_OLD = 'V' AND pos.VALUE_NEW = ' '
) chg ON bi.MANDT = chg.MANDANT AND CONCAT(bi.BUKRS, bi.BELNR, bi.GJAHR) = chg.OBJECTID
WHERE bi.DocStatus != 'V' -- Only include documents that are currently posted

UNION ALL

-- 10. Due Date Passed Without Payment
SELECT
  bi.InvoiceNumber,
  'Due Date Passed Without Payment' AS "Activity",
  TO_TIMESTAMP(ivl.DueDate, 'YYYYMMDD') AS "EventTime",
  'System' AS "ProcessorUser",
  vend.Name1 AS "VendorName",
  bi.BUKRS AS "CompanyCode",
  ivl.InvoiceGrossAmount AS "InvoiceGrossAmount",
  ivl.DueDate AS "InvoiceDueDate",
  ivl.PaymentTerms AS "PaymentTerms",
  ivl.PurchaseOrderNumber AS "PurchaseOrderNumber",
  ivl.PaymentBlockReason AS "PaymentBlockReason",
  NULL AS "Approver"
FROM BaseInvoices bi
JOIN InvoiceVendorLines ivl ON bi.MANDT = ivl.MANDT AND bi.BUKRS = ivl.BUKRS AND bi.BELNR = ivl.BELNR AND bi.GJAHR = ivl.GJAHR
LEFT JOIN LFA1 vend ON ivl.MANDT = vend.MANDT AND ivl.VendorNumber = vend.LIFNR
WHERE ivl.ClearingDoc IS NULL AND ivl.DueDate < CURRENT_DATE

UNION ALL

-- 11. Payment Proposal Created
SELECT DISTINCT
  bi.InvoiceNumber,
  'Payment Proposal Created' AS "Activity",
  TO_TIMESTAMP(CONCAT(reguh.LAUFD, '000000'), 'YYYYMMDDHH24MISS') AS "EventTime", -- Time is not available in REGUH
  reguh.USNAM AS "ProcessorUser",
  vend.Name1 AS "VendorName",
  bi.BUKRS AS "CompanyCode",
  ivl.InvoiceGrossAmount AS "InvoiceGrossAmount",
  ivl.DueDate AS "InvoiceDueDate",
  ivl.PaymentTerms AS "PaymentTerms",
  ivl.PurchaseOrderNumber AS "PurchaseOrderNumber",
  ivl.PaymentBlockReason AS "PaymentBlockReason",
  NULL AS "Approver"
FROM BaseInvoices bi
JOIN InvoiceVendorLines ivl ON bi.MANDT = ivl.MANDT AND bi.BUKRS = ivl.BUKRS AND bi.BELNR = ivl.BELNR AND bi.GJAHR = ivl.GJAHR
JOIN REGUP ON ivl.MANDT = REGUP.MANDT AND ivl.BUKRS = REGUP.BUKRS AND ivl.BELNR = REGUP.BELNR AND ivl.GJAHR = REGUP.GJAHR
JOIN REGUH ON REGUP.MANDT = REGUH.MANDT AND REGUP.LAUFD = REGUH.LAUFD AND REGUP.LAUFI = REGUH.LAUFI AND REGUP.XVORL = REGUH.XVORL AND REGUP.ZBUKR = REGUH.ZBUKR
LEFT JOIN LFA1 vend ON ivl.MANDT = vend.MANDT AND ivl.VendorNumber = vend.LIFNR

UNION ALL

-- 12. Invoice Cleared By Payment
SELECT
  bi.InvoiceNumber,
  'Invoice Cleared By Payment' AS "Activity",
  TO_TIMESTAMP(ivl.ClearingDate, 'YYYYMMDD') AS "EventTime",
  NULL AS "ProcessorUser", -- Clearing user is not on the vendor line
  vend.Name1 AS "VendorName",
  bi.BUKRS AS "CompanyCode",
  ivl.InvoiceGrossAmount AS "InvoiceGrossAmount",
  ivl.DueDate AS "InvoiceDueDate",
  ivl.PaymentTerms AS "PaymentTerms",
  ivl.PurchaseOrderNumber AS "PurchaseOrderNumber",
  ivl.PaymentBlockReason AS "PaymentBlockReason",
  NULL AS "Approver"
FROM BaseInvoices bi
JOIN InvoiceVendorLines ivl ON bi.MANDT = ivl.MANDT AND bi.BUKRS = ivl.BUKRS AND bi.BELNR = ivl.BELNR AND bi.GJAHR = ivl.GJAHR
LEFT JOIN LFA1 vend ON ivl.MANDT = vend.MANDT AND ivl.VendorNumber = vend.LIFNR
WHERE ivl.ClearingDoc IS NOT NULL

UNION ALL

-- 13. Invoice Reversed
SELECT
  bi.InvoiceNumber,
  'Invoice Reversed' AS "Activity",
  TO_TIMESTAMP(bi.ReversalDate, 'YYYYMMDD') AS "EventTime",
  NULL AS "ProcessorUser", -- Reversal user not in BKPF, would require another join
  vend.Name1 AS "VendorName",
  bi.BUKRS AS "CompanyCode",
  ivl.InvoiceGrossAmount AS "InvoiceGrossAmount",
  ivl.DueDate AS "InvoiceDueDate",
  ivl.PaymentTerms AS "PaymentTerms",
  ivl.PurchaseOrderNumber AS "PurchaseOrderNumber",
  NULL AS "PaymentBlockReason",
  NULL AS "Approver"
FROM BaseInvoices bi
JOIN InvoiceVendorLines ivl ON bi.MANDT = ivl.MANDT AND bi.BUKRS = ivl.BUKRS AND bi.BELNR = ivl.BELNR AND bi.GJAHR = ivl.GJAHR
LEFT JOIN LFA1 vend ON ivl.MANDT = vend.MANDT AND ivl.VendorNumber = vend.LIFNR
WHERE bi.ReversalDoc IS NOT NULL;
`

步骤

安装前准备：确保您的Python环境中已安装必要的库。这包括用于数据处理的 pandas 库和用于SAP RFC通信的 pyrfc 库。此外，必须从SAP支持门户下载SAP NetWeaver RFC SDK，并确保其 lib 目录在您的系统路径中可访问。
SAP连接设置：在Python脚本中，配置SAP连接字典，填入您系统的具体信息：应用服务器主机 (ashost)、系统编号 (sysnr)、客户端 (client)、用户 (user) 和密码 (passwd)。确保指定的SAP用户具有所需的权限，主要是对所需表进行 RFC_READ_TABLE 访问的 S_RFC 和 S_TABU_DIS 权限。
定义抽取参数：在脚本顶部设置数据抽取的核心参数。这包括 date_range（发票录入的开始和结束日期）以及 company_codes 列表，以便将抽取范围限定在相关业务单元。
从SAP表中抽取数据：执行Python脚本。脚本将通过RFC协议连接到SAP系统，并多次调用 RFC_READ_TABLE 函数，从核心应付账款表中抽取原始数据。这包括 BKPF（抬头）、BSEG（行项目）、RSEG（发票收据详情）、LFA1（供应商主数据）、CDHDR/CDPOS（冻结更改日志）和 BSAK（已清算项目）。
内存数据连接：脚本使用 pandas 库对抽取的DataFrame进行连接操作。它将发票抬头与行项目关联起来，通过供应商名称丰富数据，并将发票与相应的更改凭证记录连接起来。这为每张发票创建了一个全面、扁平化的数据集。
事件日志转换：这是脚本的核心逻辑。它遍历连接后的数据，为13个所需的业务活动分别创建独立的记录。对于每个活动（例如，“发票已暂存”、“付款冻结已设置”），脚本会分配正确的活动名称，并从SAP数据中相关的日期和时间字段确定精确的时间戳。
处理计算生成的事件：脚本会计算那些未直接记录在SAP中的事件。例如，“到期日未付款”事件是通过比较所有未清算发票的到期日 (ZFBDT) 与当前日期来生成的。
属性填充：对于每条事件记录，脚本会填充一组一致的属性。这包括 InvoiceNumber、Activity 和 EventTime 等必需字段，以及 VendorName、InvoiceGrossAmount 和 CompanyCode 等推荐属性，为流程分析提供丰富的上下文信息。
最终整合：脚本将所有独立的事件DataFrame收集到一个最终的事件日志DataFrame中。它确保所有记录中的列名和数据类型保持一致。
导出为CSV：最后一步是将整合后的事件日志DataFrame保存到CSV文件。此文件已格式化，可直接上传到ProcessMind等流程挖掘工具中。脚本将在运行目录中输出一个名为 sap_ap_event_log.csv 的文件。

配置

SAP连接参数：脚本需要一个包含连接详情的字典：ashost（主机名或IP地址）、sysnr（系统编号）、client（例如“100”）、user（SAP用户名）和passwd（SAP密码）。
日期范围：定义“YYYYMMDD”格式的start_date和end_date。此筛选器应用于发票录入日期（BKPF-CPUDT），以限制数据提取的范围。建议初始分析选择3-6个月的范围，以平衡数据量和洞察力。
公司代码（BUKRS）：指定要包含在数据提取中的SAP公司代码列表。这是一个关键筛选器，可确保您分析正确的组织实体（例如['1000', '2000']）。
凭证类型（BLART）：虽然此脚本中未明确将其作为参数，但您可以轻松地在对BKPF表的RFC_READ_TABLE调用中添加凭证类型筛选器（例如，KR表示供应商发票，RE表示发票收据），以进一步优化数据集。
性能与分块：脚本会根据筛选器读取完整的表。对于非常大的数据集（数百万张发票），RFC_READ_TABLE可能会超时或消耗过多内存。如果遇到性能问题，请考虑按月迭代日期范围来实现分块处理。
SAP权限：SAP用户必须具备RFC执行权限以及显示以下表数据的权限：BKPF、BSEG、RSEG、LFA1、CDHDR、CDPOS、BSAK、BSIK、REGUH、REGUP。这通常通过授权对象S_TABU_DIS授予。

a 查询示例 python

`python
import pandas as pd
from pyrfc import Connection
from datetime import datetime

class=class="hl-string">"hl-comment"># --- 1. SETTINGS & PARAMETERS ---

class=class="hl-string">"hl-comment"># SAP Connection Details (replace with your system's details)
sap_conn_params = {
    class="hl-string">"user": class="hl-string">"[SAP_USER]",
    class="hl-string">"passwd": class="hl-string">"[SAP_PASSWORD]",
    class="hl-string">"ashost": class="hl-string">"[SAP_HOST]",
    class="hl-string">"sysnr": class="hl-string">"[SYSTEM_NUMBER]",
    class="hl-string">"client": class="hl-string">"[CLIENT]"
}

class=class="hl-string">"hl-comment"># Extraction Parameters
company_codes = [class="hl-string">'[YOUR_COMPANY_CODE_1]', class="hl-string">'[YOUR_COMPANY_CODE_2]']
date_range = {
    class="hl-string">"start_date": class="hl-string">"20230101", class=class="hl-string">"hl-comment"># YYYYMMDD
    class="hl-string">"end_date": class="hl-string">"20231231"    class=class="hl-string">"hl-comment"># YYYYMMDD
}

class=class="hl-string">"hl-comment"># --- 2. HELPER FUNCTION TO READ SAP TABLE ---
def read_sap_table(conn, table_name, fields=[], options=[], row_count=0):
    class="hl-string">"""Reads an SAP table using RFC_READ_TABLE and returns a pandas DataFrame."""
    try:
        result = conn.call(class="hl-string">'RFC_READ_TABLE',
                             QUERY_TABLE=table_name,
                             DELIMITER=class="hl-string">'|',
                             FIELDS=fields,
                             OPTIONS=options,
                             ROWCOUNT=row_count)
        
        data = result[class="hl-string">'DATA']
        if not data:
            return pd.DataFrame(columns=[f[class="hl-string">'FIELDNAME'] for f in result[class="hl-string">'FIELDS']])

        df = pd.DataFrame([row[class="hl-string">'WA'].split(class="hl-string">'|') for row in data])
        df.columns = [f[class="hl-string">'FIELDNAME'] for f in result[class="hl-string">'FIELDS']]
        return df
    except Exception as e:
        print(fclass="hl-string">"Error reading table {table_name}: {e}")
        return pd.DataFrame(columns=[f[class="hl-string">'FIELDNAME'] for f in fields])

class=class="hl-string">"hl-comment"># --- 3. DATA EXTRACTION ---
print(class="hl-string">"Starting SAP data extraction...")

with Connection(**sap_conn_params) as conn:
    class=class="hl-string">"hl-comment"># Define filters
    company_code_filter = class="hl-string">" OR ".join([fclass="hl-string">"BUKRS = '{cc}'" for cc in company_codes])
    date_filter = fclass="hl-string">"CPUDT >= '{date_range['start_date']}' AND CPUDT <= '{date_range['end_date']}'"
    bkpf_filter = [fclass="hl-string">"({company_code_filter}) AND ({date_filter})"]

    class=class="hl-string">"hl-comment"># Extract main tables
    df_bkpf = read_sap_table(conn, class="hl-string">'BKPF', fields=[class="hl-string">'MANDT', class="hl-string">'BUKRS', class="hl-string">'BELNR', class="hl-string">'GJAHR', class="hl-string">'BLART', class="hl-string">'BUDAT', class="hl-string">'CPUDT', class="hl-string">'CPUTM', class="hl-string">'USNAM', class="hl-string">'TCODE', class="hl-string">'STBLG', class="hl-string">'BSTAT'], options=bkpf_filter)
    df_bkpf[class="hl-string">'AWKEY'] = df_bkpf[class="hl-string">'BELNR'] + df_bkpf[class="hl-string">'GJAHR']
    df_bkpf[class="hl-string">'OBJECTKEY'] = df_bkpf[class="hl-string">'MANDT'] + df_bkpf[class="hl-string">'BUKRS'] + df_bkpf[class="hl-string">'BELNR'] + df_bkpf[class="hl-string">'GJAHR']
    doc_keys = df_bkpf[[class="hl-string">'BELNR', class="hl-string">'GJAHR', class="hl-string">'BUKRS']].drop_duplicates().to_dict(class="hl-string">'records')
    bseg_filter = class="hl-string">" OR ".join([fclass="hl-string">"(BELNR = '{k['BELNR']}' AND GJAHR = '{k['GJAHR']}' AND BUKRS = '{k['BUKRS']}')" for k in doc_keys])
    df_bseg = read_sap_table(conn, class="hl-string">'BSEG', fields=[class="hl-string">'BELNR', class="hl-string">'GJAHR', class="hl-string">'BUKRS', class="hl-string">'BUZEI', class="hl-string">'LIFNR', class="hl-string">'ZFBDT', class="hl-string">'ZTERM', class="hl-string">'ZLSPR', class="hl-string">'WRBTR'])

    class=class="hl-string">"hl-comment"># Extract more tables based on findings
    rseg_filter = [fclass="hl-string">"BELNR IN ({','.join(df_bkpf['BELNR'].unique())}) AND GJAHR IN ({','.join(df_bkpf['GJAHR'].unique())})"]
    df_rseg = read_sap_table(conn, class="hl-string">'RSEG', fields=[class="hl-string">'BELNR', class="hl-string">'GJAHR', class="hl-string">'EBELN', class="hl-string">'EBELP', class="hl-string">'LFBNR'])
    
    vendor_numbers = df_bseg[class="hl-string">'LIFNR'].dropna().unique()
    lfa1_filter = [fclass="hl-string">"LIFNR IN ({','.join(vendor_numbers)})"]
    df_lfa1 = read_sap_table(conn, class="hl-string">'LFA1', fields=[class="hl-string">'LIFNR', class="hl-string">'NAME1'])

    class=class="hl-string">"hl-comment"># Change Documents (CDHDR/CDPOS) for Payment Blocks
    cdhdr_filter = [fclass="hl-string">"OBJECTCLAS = 'BELEGV' AND OBJECTID IN ({','.join(df_bkpf['OBJECTKEY'].unique())}) AND TCODE = 'FB02'"]
    df_cdhdr = read_sap_table(conn, class="hl-string">'CDHDR', fields=[class="hl-string">'OBJECTCLAS', class="hl-string">'OBJECTID', class="hl-string">'CHANGENR', class="hl-string">'USERNAME', class="hl-string">'UDATE', class="hl-string">'UTIME'])
    cdpos_filter = [fclass="hl-string">"OBJECTCLAS = 'BELEGV' AND FNAME = 'ZLSPR' AND TABNAME = 'BSEG' AND CHANGENR IN ({','.join(df_cdhdr['CHANGENR'].unique())})"]
    df_cdpos = read_sap_table(conn, class="hl-string">'CDPOS', fields=[class="hl-string">'CHANGENR', class="hl-string">'TABKEY', class="hl-string">'FNAME', class="hl-string">'VALUE_NEW', class="hl-string">'VALUE_OLD'])
    
    class=class="hl-string">"hl-comment"># Cleared Items
    bsak_filter = class="hl-string">" OR ".join([fclass="hl-string">"(BELNR = '{k['BELNR']}' AND GJAHR = '{k['GJAHR']}' AND BUKRS = '{k['BUKRS']}')" for k in doc_keys])
    df_bsak = read_sap_table(conn, class="hl-string">'BSAK', fields=[class="hl-string">'BELNR', class="hl-string">'GJAHR', class="hl-string">'BUKRS', class="hl-string">'AUGDT', class="hl-string">'AUGBL'])

    class=class="hl-string">"hl-comment"># Payment Proposal
    regup_filter = class="hl-string">" OR ".join([fclass="hl-string">"(VBLNR = '{k['BELNR']}' AND GJAHR = '{k['GJAHR']}' AND BUKRS = '{k['BUKRS']}')" for k in doc_keys])
    df_regup = read_sap_table(conn, class="hl-string">'REGUP', fields=[class="hl-string">'LAUFD', class="hl-string">'LAUFI', class="hl-string">'VBLNR', class="hl-string">'GJAHR', class="hl-string">'BUKRS'])
    reguh_filter = [fclass="hl-string">"LAUFD IN ({','.join(df_regup['LAUFD'].unique())}) AND LAUFI IN ({','.join(df_regup['LAUFI'].unique())})"]
    df_reguh = read_sap_table(conn, class="hl-string">'REGUH', fields=[class="hl-string">'LAUFD', class="hl-string">'LAUFI', class="hl-string">'VBLNR'])


print(class="hl-string">"Extraction complete. Starting data transformation...")

class=class="hl-string">"hl-comment"># --- 4. DATA TRANSFORMATION & JOINING ---
class=class="hl-string">"hl-comment"># Merge core invoice data
df_inv = pd.merge(df_bkpf, df_bseg.loc[df_bseg[class="hl-string">'LIFNR'].notna()], on=[class="hl-string">'BELNR', class="hl-string">'GJAHR', class="hl-string">'BUKRS'])
df_inv = pd.merge(df_inv, df_lfa1, on=class="hl-string">'LIFNR', how=class="hl-string">'left')
df_inv = pd.merge(df_inv, df_rseg, on=[class="hl-string">'BELNR', class="hl-string">'GJAHR'], how=class="hl-string">'left')

class=class="hl-string">"hl-comment"># --- 5. EVENT LOG CREATION ---
all_events = []

class=class="hl-string">"hl-comment"># Invoice Document Created
df_created = df_inv.copy()
df_created[class="hl-string">'Activity'] = class="hl-string">'Invoice Document Created'
df_created[class="hl-string">'EventTime'] = pd.to_datetime(df_created[class="hl-string">'CPUDT'] + df_created[class="hl-string">'CPUTM'], format=class="hl-string">'%Y%m%d%H%M%S')
all_events.append(df_created)

class=class="hl-string">"hl-comment"># Invoice Parked
df_parked = df_inv[df_inv[class="hl-string">'BSTAT'] == class="hl-string">'V'].copy()
df_parked[class="hl-string">'Activity'] = class="hl-string">'Invoice Parked'
df_parked[class="hl-string">'EventTime'] = pd.to_datetime(df_parked[class="hl-string">'CPUDT'] + df_parked[class="hl-string">'CPUTM'], format=class="hl-string">'%Y%m%d%H%M%S')
all_events.append(df_parked)

class=class="hl-string">"hl-comment"># Purchase Order Matched
df_po_match = df_inv[df_inv[class="hl-string">'EBELN'].notna()].copy()
df_po_match[class="hl-string">'Activity'] = class="hl-string">'Purchase Order Matched'
df_po_match[class="hl-string">'EventTime'] = pd.to_datetime(df_po_match[class="hl-string">'CPUDT'] + df_po_match[class="hl-string">'CPUTM'], format=class="hl-string">'%Y%m%d%H%M%S')
all_events.append(df_po_match)

class=class="hl-string">"hl-comment"># Goods Receipt Matched
df_gr_match = df_inv[df_inv[class="hl-string">'LFBNR'].notna()].copy()
df_gr_match[class="hl-string">'Activity'] = class="hl-string">'Goods Receipt Matched'
df_gr_match[class="hl-string">'EventTime'] = pd.to_datetime(df_gr_match[class="hl-string">'CPUDT'] + df_gr_match[class="hl-string">'CPUTM'], format=class="hl-string">'%Y%m%d%H%M%S')
all_events.append(df_gr_match)

class=class="hl-string">"hl-comment"># Payment Block Set & Removed (from change logs)
if not df_cdhdr.empty and not df_cdpos.empty:
    df_changes = pd.merge(df_cdhdr, df_cdpos, on=class="hl-string">'CHANGENR')
    df_changes[class="hl-string">'BELNR'] = df_changes[class="hl-string">'OBJECTID'].str[7:17]
    df_changes[class="hl-string">'GJAHR'] = df_changes[class="hl-string">'OBJECTID'].str[17:21]
    df_changes[class="hl-string">'BUKRS'] = df_changes[class="hl-string">'OBJECTID'].str[3:7]
    df_changes = pd.merge(df_changes, df_inv[[class="hl-string">'BELNR', class="hl-string">'GJAHR', class="hl-string">'BUKRS']].drop_duplicates(), on=[class="hl-string">'BELNR', class="hl-string">'GJAHR', class="hl-string">'BUKRS'])
    
    df_block_set = df_changes[df_changes[class="hl-string">'VALUE_NEW'].str.strip() != class="hl-string">''].copy()
    df_block_set[class="hl-string">'Activity'] = class="hl-string">'Payment Block Set'
    df_block_set[class="hl-string">'EventTime'] = pd.to_datetime(df_block_set[class="hl-string">'UDATE'] + df_block_set[class="hl-string">'UTIME'], format=class="hl-string">'%Y%m%d%H%M%S')
    df_block_set.rename(columns={class="hl-string">'VALUE_NEW': class="hl-string">'PaymentBlockReason', class="hl-string">'USERNAME': class="hl-string">'ProcessorUser'}, inplace=True)
    all_events.append(df_block_set)

    df_block_removed = df_changes[df_changes[class="hl-string">'VALUE_OLD'].str.strip() != class="hl-string">''].copy()
    df_block_removed[class="hl-string">'Activity'] = class="hl-string">'Payment Block Removed'
    df_block_removed[class="hl-string">'EventTime'] = pd.to_datetime(df_block_removed[class="hl-string">'UDATE'] + df_block_removed[class="hl-string">'UTIME'], format=class="hl-string">'%Y%m%d%H%M%S')
    df_block_removed.rename(columns={class="hl-string">'VALUE_OLD': class="hl-string">'PaymentBlockReason', class="hl-string">'USERNAME': class="hl-string">'ProcessorUser'}, inplace=True)
    all_events.append(df_block_removed)

    class=class="hl-string">"hl-comment"># Invoice Routed For Approval & Approved (inferred from blocks)
    class=class="hl-string">"hl-comment"># Assumption: A specific block reason, e.g., class="hl-string">'A', signifies approval workflow.
    df_routed = df_block_set[df_block_set[class="hl-string">'PaymentBlockReason'].str.strip() == class="hl-string">'[A]'].copy()
    df_routed[class="hl-string">'Activity'] = class="hl-string">'Invoice Routed For Approval'
    all_events.append(df_routed)

    df_approved = df_block_removed[df_block_removed[class="hl-string">'PaymentBlockReason'].str.strip() == class="hl-string">'[A]'].copy()
    df_approved[class="hl-string">'Activity'] = class="hl-string">'Invoice Approved'
    df_approved[class="hl-string">'Approver'] = df_approved[class="hl-string">'ProcessorUser']
    all_events.append(df_approved)

class=class="hl-string">"hl-comment"># Invoice Posted
df_posted = df_inv[df_inv[class="hl-string">'BSTAT'] != class="hl-string">'V'].copy()
df_posted[class="hl-string">'Activity'] = class="hl-string">'Invoice Posted'
df_posted[class="hl-string">'EventTime'] = pd.to_datetime(df_posted[class="hl-string">'BUDAT'], format=class="hl-string">'%Y%m%d')
all_events.append(df_posted)

class=class="hl-string">"hl-comment"># Invoice Cleared By Payment
if not df_bsak.empty:
    df_cleared = pd.merge(df_inv, df_bsak[df_bsak[class="hl-string">'AUGDT'] != class="hl-string">'00000000'], on=[class="hl-string">'BELNR', class="hl-string">'GJAHR', class="hl-string">'BUKRS'])
    df_cleared[class="hl-string">'Activity'] = class="hl-string">'Invoice Cleared By Payment'
    df_cleared[class="hl-string">'EventTime'] = pd.to_datetime(df_cleared[class="hl-string">'AUGDT'], format=class="hl-string">'%Y%m%d')
    all_events.append(df_cleared)

class=class="hl-string">"hl-comment"># Invoice Reversed
df_reversed = df_inv[df_inv[class="hl-string">'STBLG'].notna() & (df_inv[class="hl-string">'STBLG'] != class="hl-string">'')].copy()
if not df_reversed.empty:
    reversal_docs = pd.merge(df_reversed, df_bkpf[[class="hl-string">'BELNR', class="hl-string">'BUDAT']], left_on=class="hl-string">'STBLG', right_on=class="hl-string">'BELNR', suffixes=(class="hl-string">'', class="hl-string">'_rev'))
    reversal_docs[class="hl-string">'Activity'] = class="hl-string">'Invoice Reversed'
    reversal_docs[class="hl-string">'EventTime'] = pd.to_datetime(reversal_docs[class="hl-string">'BUDAT_rev'], format=class="hl-string">'%Y%m%d')
    all_events.append(reversal_docs)

class=class="hl-string">"hl-comment"># Payment Proposal Created
if not df_regup.empty and not df_reguh.empty:
    df_proposal = pd.merge(df_regup, df_reguh, on=[class="hl-string">'LAUFD', class="hl-string">'LAUFI'])
    df_proposal = pd.merge(df_proposal, df_inv, left_on=[class="hl-string">'VBLNR', class="hl-string">'GJAHR', class="hl-string">'BUKRS'], right_on=[class="hl-string">'BELNR', class="hl-string">'GJAHR', class="hl-string">'BUKRS'])
    df_proposal[class="hl-string">'Activity'] = class="hl-string">'Payment Proposal Created'
    df_proposal[class="hl-string">'EventTime'] = pd.to_datetime(df_proposal[class="hl-string">'LAUFD'], format=class="hl-string">'%Y%m%d')
    all_events.append(df_proposal)

class=class="hl-string">"hl-comment"># Due Date Passed Without Payment (Calculated)
if not df_bsak.empty:
    cleared_keys = df_bsak[class="hl-string">'BELNR'] + df_bsak[class="hl-string">'GJAHR'] + df_bsak[class="hl-string">'BUKRS']
    df_open = df_inv[~(df_inv[class="hl-string">'BELNR'] + df_inv[class="hl-string">'GJAHR'] + df_inv[class="hl-string">'BUKRS']).isin(cleared_keys)].copy()
    df_open[class="hl-string">'DueDate'] = pd.to_datetime(df_open[class="hl-string">'ZFBDT'], format=class="hl-string">'%Y%m%d', errors=class="hl-string">'coerce')
    df_overdue = df_open[(df_open[class="hl-string">'DueDate'] < datetime.now()) & (df_open[class="hl-string">'DueDate'].notna())].copy()
    if not df_overdue.empty:
        df_overdue[class="hl-string">'Activity'] = class="hl-string">'Due Date Passed Without Payment'
        df_overdue[class="hl-string">'EventTime'] = df_overdue[class="hl-string">'DueDate']
        all_events.append(df_overdue)

class=class="hl-string">"hl-comment"># --- 6. FINAL ASSEMBLY & EXPORT ---
print(class="hl-string">"Assembling final event log...")

df_event_log = pd.concat(all_events, ignore_index=True, sort=False)

class=class="hl-string">"hl-comment"># Rename and select columns
column_mapping = {
    class="hl-string">'BELNR': class="hl-string">'InvoiceNumber',
    class="hl-string">'USNAM': class="hl-string">'ProcessorUser',
    class="hl-string">'NAME1': class="hl-string">'VendorName',
    class="hl-string">'BUKRS': class="hl-string">'CompanyCode',
    class="hl-string">'WRBTR': class="hl-string">'InvoiceGrossAmount',
    class="hl-string">'ZFBDT': class="hl-string">'InvoiceDueDate',
    class="hl-string">'ZTERM': class="hl-string">'PaymentTerms',
    class="hl-string">'EBELN': class="hl-string">'PurchaseOrderNumber',
    class="hl-string">'ZLSPR': class="hl-string">'PaymentBlockReason'
}
df_event_log.rename(columns=column_mapping, inplace=True)

class=class="hl-string">"hl-comment"># Ensure required columns exist
final_cols = [
    class="hl-string">'InvoiceNumber', class="hl-string">'Activity', class="hl-string">'EventTime', class="hl-string">'ProcessorUser', class="hl-string">'VendorName', 
    class="hl-string">'CompanyCode', class="hl-string">'InvoiceGrossAmount', class="hl-string">'InvoiceDueDate', class="hl-string">'PaymentTerms', 
    class="hl-string">'PurchaseOrderNumber', class="hl-string">'PaymentBlockReason', class="hl-string">'Approver'
]

for col in final_cols:
    if col not in df_event_log.columns:
        df_event_log[col] = None

df_final = df_event_log[final_cols].sort_values(by=[class="hl-string">'InvoiceNumber', class="hl-string">'EventTime']).reset_index(drop=True)

class=class="hl-string">"hl-comment"># Export to CSV
df_final.to_csv(class="hl-string">'sap_ap_event_log.csv', index=False)
print(fclass="hl-string">"Event log created successfully with {len(df_final)} events.")

`

步骤

准备工作及配置：安装经认证的第三方ETL工具（例如 Theobald Xtract Universal、Talend with SAP Connector）。确保您拥有有效许可，并能与SAP ECC系统建立网络连接。配置一个SAP用户账户，并赋予其必要的RFC和表读取权限（例如，对所需表有 S_RFC 和 S_TABU_DIS 的访问权限）。
创建SAP连接：在ETL工具中，创建一个新的SAP连接。提供应用服务器详情、系统ID、客户端编号以及该专用SAP用户账户的凭据。测试连接以确保成功。
定义基础数据源：将以下核心SAP表添加为抽取作业的数据源：BKPF（会计凭证抬头）、BSEG（会计凭证行项目）、LFA1（供应商主数据）、RBKP（MM发票抬头）、RSEG（MM发票行项目）、CDHDR（更改凭证抬头）和 CDPOS（更改凭证项目）。
建立表连接：通过连接这些表来创建一个基础视图或数据流。将 BKPF 与 BSEG 通过 BUKRS、BELNR 和 GJAHR 进行连接。将 BKPF 与 LFA1 通过 LIFNR 进行连接。将 BSEG 与 RSEG 基于 MM 凭证信息进行连接（将 BELNR、GJAHR 映射到 RSEG.BELNR、RSEG.GJAHR）。将 BKPF 与 CDHDR 通过 OBJECTCLAS='BELEG' 和 OBJECTID = CompanyCode + DocumentNumber + FiscalYear 进行连接。
应用全局筛选器：在源级别应用筛选器以限制数据范围。常见筛选条件包括 BKPF.BUKRS（公司代码）、BKPF.GJAHR（会计年度）和 BKPF.BLART（凭证类型，例如 'KR'、'RE'）。对 BKPF.CPUDT（录入日期）应用日期范围筛选器对于控制数据量至关重要。
构建“发票凭证已创建”活动：创建第一个数据流。将 Activity 设置为“发票凭证已创建”。 EventTime 是由 BKPF.CPUDT（日期）和 BKPF.CPUTM（时间）组合而成。从连接后的表中选择所有必需和推荐的属性。
构建后续活动流：对于其余的12项活动，创建单独的数据流或转换路径。应用特定逻辑和筛选器，以隔离该事件的记录。例如，对于“发票已暂存”活动，筛选 BKPF.BSTAT = 'V'。对于基于更改的事件，筛选 CDHDR 和 CDPOS 以查找特定的表、字段和值更改（例如，FNAME = 'ZLSPL' 表示付款冻结）。
隔离更改凭证事件：对于“付款冻结已设置”和“付款冻结已移除”等活动，使用 CDHDR 和 CDPOS 中的数据。 EventTime 将是 CDHDR.UDATE 和 CDHDR.UTIME。 ProcessorUser 是 CDHDR.USERNAME。
计算“到期日已过”事件：创建一个专门的转换步骤来计算此活动。逻辑为：IF BSEG.AUGBL IS NULL AND CURRENT_DATE > BSEG.ZFBDT THEN create event。此活动的 EventTime 应设置为到期日（BSEG.ZFBDT）。
合并所有流：在您的ETL工具中使用“联合（Union）”或“合并（Merge）”转换，将所有13个独立的活动流合并成一个数据集。确保在合并之前，所有数据流中的列名、顺序和数据类型保持一致。
最终格式化与导出：审阅最终合并的数据集。确保 EventTime 列采用标准的时间戳格式（例如，YYYY-MM-DD HH:MI:SS）。将最终事件日志导出为UTF-8编码的CSV文件，即可上传至您的流程挖掘工具。

配置

ETL工具及连接器：需要有许可的ETL工具并配备经认证的SAP连接器。此方法依赖于工具的图形界面进行配置，而非直接编写ABAP或SQL代码。
SAP用户权限：SAP用户需要RFC执行权限以及对以下表的显示权限：BKPF、BSEG、LFA1、T001、T003T、RBKP、RSEG、CDHDR、CDPOS、REGUH、REGUP 和 LFB1。
日期范围筛选：强烈建议在初始加载时使用如BKPF.CPUDT之类的录入日期字段进行筛选。首次分析通常选择6-12个月的范围。后续加载可基于更改日期采用增量机制。
关键来源筛选器：
- BKPF.BUKRS（公司代码）：选择特定的公司代码以聚焦分析。
- BKPF.BLART（凭证类型）：筛选相关的发票类型，通常是KR（供应商发票）和RE（MM发票收据）。
性能考量：从更改凭证表（CDHDR、CDPOS）中提取数据可能消耗大量资源。请对OBJECTCLAS = 'BELEG'和相关事务代码（TCODE）应用特定筛选器，以最大程度地降低性能影响。避免全表扫描。

a 查询示例 config

`config
-- This is a logical representation of the ETL configuration.
-- Each 'SELECT' block represents a separate data stream for an activity,
-- which are then combined using a UNION ALL transformation in the ETL tool.

-- Base Data Source: A join of BKPF, BSEG, LFA1, RBKP, RSEG
-- Let's call this joined view 'INVOICE_DETAILS'

-- 1. Invoice Document Created
SELECT
    CONCAT(B.BUKRS, B.BELNR, B.GJAHR) AS InvoiceNumber, -- Case ID
    'Invoice Document Created' AS Activity,
    TO_TIMESTAMP(B.CPUDT, B.CPUTM) AS EventTime,
    B.USNAM AS ProcessorUser,
    L.NAME1 AS VendorName,
    B.BUKRS AS CompanyCode,
    B.WAERS AS Currency,
    (SELECT SUM(WRBTR) FROM BSEG S WHERE S.BELNR = B.BELNR AND S.GJAHR = B.GJAHR AND S.BUKRS = B.BUKRS) AS InvoiceGrossAmount, 
    S1.ZFBDT AS InvoiceDueDate, -- Due date from first line item
    S1.ZTERM AS PaymentTerms,
    S1.EBELN AS PurchaseOrderNumber,
    S1.ZLSPR AS PaymentBlockReason,
    '[Workflow Approver]' AS Approver
FROM BKPF AS B
JOIN BSEG AS S1 ON B.BELNR = S1.BELNR AND B.GJAHR = S1.GJAHR AND B.BUKRS = S1.BUKRS AND S1.KOART = 'K' -- First vendor line item
JOIN LFA1 AS L ON B.LIFNR = L.LIFNR
WHERE B.BLART IN ('KR', 'RE') AND B.BUKRS = '[YourCompanyCode]' AND B.CPUDT BETWEEN '[StartDate]' AND '[EndDate]'

UNION ALL

-- 2. Invoice Parked
SELECT
    CONCAT(B.BUKRS, B.BELNR, B.GJAHR) AS InvoiceNumber,
    'Invoice Parked' AS Activity,
    TO_TIMESTAMP(B.CPUDT, B.CPUTM) AS EventTime,
    B.USNAM AS ProcessorUser, L.NAME1 AS VendorName, B.BUKRS AS CompanyCode, B.WAERS AS Currency, (SELECT SUM(WRBTR) FROM BSEG S WHERE S.BELNR = B.BELNR AND S.GJAHR = B.GJAHR AND S.BUKRS = B.BUKRS) AS InvoiceGrossAmount, S1.ZFBDT AS InvoiceDueDate, S1.ZTERM AS PaymentTerms, S1.EBELN AS PurchaseOrderNumber, S1.ZLSPR AS PaymentBlockReason, '[Workflow Approver]' AS Approver
FROM BKPF AS B
JOIN BSEG AS S1 ON B.BELNR = S1.BELNR AND B.GJAHR = S1.GJAHR AND B.BUKRS = S1.BUKRS AND S1.KOART = 'K'
JOIN LFA1 AS L ON B.LIFNR = L.LIFNR
WHERE B.BSTAT = 'V' AND B.BLART IN ('KR', 'RE') AND B.BUKRS = '[YourCompanyCode]' AND B.CPUDT BETWEEN '[StartDate]' AND '[EndDate]'

UNION ALL

-- 3. Purchase Order Matched
SELECT DISTINCT
    CONCAT(B.BUKRS, B.BELNR, B.GJAHR) AS InvoiceNumber,
    'Purchase Order Matched' AS Activity, 
    TO_TIMESTAMP(B.CPUDT, B.CPUTM) AS EventTime, 
    B.USNAM AS ProcessorUser, L.NAME1 AS VendorName, B.BUKRS AS CompanyCode, B.WAERS AS Currency, (SELECT SUM(WRBTR) FROM BSEG S WHERE S.BELNR = B.BELNR AND S.GJAHR = B.GJAHR AND S.BUKRS = B.BUKRS) AS InvoiceGrossAmount, S1.ZFBDT AS InvoiceDueDate, S1.ZTERM AS PaymentTerms, S1.EBELN AS PurchaseOrderNumber, S1.ZLSPR AS PaymentBlockReason, '[Workflow Approver]' AS Approver
FROM BKPF AS B
JOIN BSEG AS S1 ON B.BELNR = S1.BELNR AND B.GJAHR = S1.GJAHR AND B.BUKRS = S1.BUKRS AND S1.KOART = 'K'
JOIN LFA1 AS L ON B.LIFNR = L.LIFNR
WHERE S1.EBELN IS NOT NULL AND B.BLART IN ('KR', 'RE') AND B.BUKRS = '[YourCompanyCode]' AND B.CPUDT BETWEEN '[StartDate]' AND '[EndDate]'

UNION ALL

-- 4. Goods Receipt Matched
SELECT DISTINCT
    CONCAT(B.BUKRS, B.BELNR, B.GJAHR) AS InvoiceNumber,
    'Goods Receipt Matched' AS Activity, 
    TO_TIMESTAMP(B.CPUDT, B.CPUTM) AS EventTime, 
    B.USNAM AS ProcessorUser, L.NAME1 AS VendorName, B.BUKRS AS CompanyCode, B.WAERS AS Currency, (SELECT SUM(WRBTR) FROM BSEG S WHERE S.BELNR = B.BELNR AND S.GJAHR = B.GJAHR AND S.BUKRS = B.BUKRS) AS InvoiceGrossAmount, S1.ZFBDT AS InvoiceDueDate, S1.ZTERM AS PaymentTerms, S1.EBELN AS PurchaseOrderNumber, S1.ZLSPR AS PaymentBlockReason, '[Workflow Approver]' AS Approver
FROM BKPF AS B
JOIN BSEG AS S1 ON B.BELNR = S1.BELNR AND B.GJAHR = S1.GJAHR AND B.BUKRS = S1.BUKRS AND S1.KOART = 'K'
JOIN RSEG AS R ON S1.BELNR = R.BELNR AND S1.GJAHR = R.GJAHR
JOIN LFA1 AS L ON B.LIFNR = L.LIFNR
WHERE R.LFBNR IS NOT NULL AND B.BLART IN ('KR', 'RE') AND B.BUKRS = '[YourCompanyCode]' AND B.CPUDT BETWEEN '[StartDate]' AND '[EndDate]'

UNION ALL

-- 5. Payment Block Set
SELECT
    CONCAT(SUBSTRING(H.OBJECTID, 1, 4), SUBSTRING(H.OBJECTID, 5, 10), SUBSTRING(H.OBJECTID, 15, 4)) AS InvoiceNumber,
    'Payment Block Set' AS Activity,
    TO_TIMESTAMP(H.UDATE, H.UTIME) AS EventTime,
    H.USERNAME AS ProcessorUser, L.NAME1 AS VendorName, B.BUKRS AS CompanyCode, B.WAERS AS Currency, (SELECT SUM(WRBTR) FROM BSEG S WHERE S.BELNR = B.BELNR AND S.GJAHR = B.GJAHR AND S.BUKRS = B.BUKRS) AS InvoiceGrossAmount, S1.ZFBDT AS InvoiceDueDate, S1.ZTERM AS PaymentTerms, S1.EBELN AS PurchaseOrderNumber, P.VALUE_NEW AS PaymentBlockReason, '[Workflow Approver]' AS Approver
FROM CDHDR H
JOIN CDPOS P ON H.CHANGENR = P.CHANGENR
JOIN BKPF B ON SUBSTRING(H.OBJECTID, 5, 10) = B.BELNR AND SUBSTRING(H.OBJECTID, 1, 4) = B.BUKRS AND SUBSTRING(H.OBJECTID, 15, 4) = B.GJAHR
JOIN BSEG AS S1 ON B.BELNR = S1.BELNR AND B.GJAHR = S1.GJAHR AND B.BUKRS = S1.BUKRS AND S1.KOART = 'K'
JOIN LFA1 AS L ON B.LIFNR = L.LIFNR
WHERE H.OBJECTCLAS = 'BELEG' AND P.TABNAME = 'BSEG' AND P.FNAME = 'ZLSPR' AND P.VALUE_NEW <> ''
AND B.BUKRS = '[YourCompanyCode]' AND H.UDATE BETWEEN '[StartDate]' AND '[EndDate]'

UNION ALL

-- 6. Invoice Routed For Approval (Inferred from specific block)
SELECT
    CONCAT(B.BUKRS, B.BELNR, B.GJAHR) AS InvoiceNumber, 'Invoice Routed For Approval' AS Activity, TO_TIMESTAMP(H.UDATE, H.UTIME) AS EventTime, H.USERNAME AS ProcessorUser, L.NAME1 AS VendorName, B.BUKRS AS CompanyCode, B.WAERS AS Currency, (SELECT SUM(WRBTR) FROM BSEG S WHERE S.BELNR = B.BELNR AND S.GJAHR = B.GJAHR AND S.BUKRS = B.BUKRS) AS InvoiceGrossAmount, S1.ZFBDT AS InvoiceDueDate, S1.ZTERM AS PaymentTerms, S1.EBELN AS PurchaseOrderNumber, P.VALUE_NEW AS PaymentBlockReason, '[Workflow Approver]' AS Approver
FROM CDHDR H JOIN CDPOS P ON H.CHANGENR = P.CHANGENR JOIN BKPF B ON SUBSTRING(H.OBJECTID, 5, 10) = B.BELNR AND SUBSTRING(H.OBJECTID, 1, 4) = B.BUKRS AND SUBSTRING(H.OBJECTID, 15, 4) = B.GJAHR JOIN BSEG AS S1 ON B.BELNR = S1.BELNR AND B.GJAHR = S1.GJAHR AND B.BUKRS = S1.BUKRS AND S1.KOART = 'K' JOIN LFA1 AS L ON B.LIFNR = L.LIFNR
WHERE H.OBJECTCLAS = 'BELEG' AND P.TABNAME = 'BSEG' AND P.FNAME = 'ZLSPR' AND P.VALUE_NEW = '[ApprovalBlockReasonCode]'
AND B.BUKRS = '[YourCompanyCode]' AND H.UDATE BETWEEN '[StartDate]' AND '[EndDate]'

UNION ALL

-- 7. Invoice Approved (Inferred from removal of specific block)
SELECT
    CONCAT(B.BUKRS, B.BELNR, B.GJAHR) AS InvoiceNumber, 'Invoice Approved' AS Activity, TO_TIMESTAMP(H.UDATE, H.UTIME) AS EventTime, H.USERNAME AS ProcessorUser, L.NAME1 AS VendorName, B.BUKRS AS CompanyCode, B.WAERS AS Currency, (SELECT SUM(WRBTR) FROM BSEG S WHERE S.BELNR = B.BELNR AND S.GJAHR = B.GJAHR AND S.BUKRS = B.BUKRS) AS InvoiceGrossAmount, S1.ZFBDT AS InvoiceDueDate, S1.ZTERM AS PaymentTerms, S1.EBELN AS PurchaseOrderNumber, '' AS PaymentBlockReason, H.USERNAME AS Approver
FROM CDHDR H JOIN CDPOS P ON H.CHANGENR = P.CHANGENR JOIN BKPF B ON SUBSTRING(H.OBJECTID, 5, 10) = B.BELNR AND SUBSTRING(H.OBJECTID, 1, 4) = B.BUKRS AND SUBSTRING(H.OBJECTID, 15, 4) = B.GJAHR JOIN BSEG AS S1 ON B.BELNR = S1.BELNR AND B.GJAHR = S1.GJAHR AND B.BUKRS = S1.BUKRS AND S1.KOART = 'K' JOIN LFA1 AS L ON B.LIFNR = L.LIFNR
WHERE H.OBJECTCLAS = 'BELEG' AND P.TABNAME = 'BSEG' AND P.FNAME = 'ZLSPR' AND P.VALUE_OLD = '[ApprovalBlockReasonCode]' AND P.VALUE_NEW = ''
AND B.BUKRS = '[YourCompanyCode]' AND H.UDATE BETWEEN '[StartDate]' AND '[EndDate]'

UNION ALL

-- 8. Payment Block Removed
SELECT
    CONCAT(SUBSTRING(H.OBJECTID, 1, 4), SUBSTRING(H.OBJECTID, 5, 10), SUBSTRING(H.OBJECTID, 15, 4)) AS InvoiceNumber, 'Payment Block Removed' AS Activity, TO_TIMESTAMP(H.UDATE, H.UTIME) AS EventTime, H.USERNAME AS ProcessorUser, L.NAME1 AS VendorName, B.BUKRS AS CompanyCode, B.WAERS AS Currency, (SELECT SUM(WRBTR) FROM BSEG S WHERE S.BELNR = B.BELNR AND S.GJAHR = B.GJAHR AND S.BUKRS = B.BUKRS) AS InvoiceGrossAmount, S1.ZFBDT AS InvoiceDueDate, S1.ZTERM AS PaymentTerms, S1.EBELN AS PurchaseOrderNumber, '' AS PaymentBlockReason, '[Workflow Approver]' AS Approver
FROM CDHDR H JOIN CDPOS P ON H.CHANGENR = P.CHANGENR JOIN BKPF B ON SUBSTRING(H.OBJECTID, 5, 10) = B.BELNR AND SUBSTRING(H.OBJECTID, 1, 4) = B.BUKRS AND SUBSTRING(H.OBJECTID, 15, 4) = B.GJAHR JOIN BSEG AS S1 ON B.BELNR = S1.BELNR AND B.GJAHR = S1.GJAHR AND B.BUKRS = S1.BUKRS AND S1.KOART = 'K' JOIN LFA1 AS L ON B.LIFNR = L.LIFNR
WHERE H.OBJECTCLAS = 'BELEG' AND P.TABNAME = 'BSEG' AND P.FNAME = 'ZLSPR' AND P.VALUE_NEW = '' AND P.VALUE_OLD <> ''
AND B.BUKRS = '[YourCompanyCode]' AND H.UDATE BETWEEN '[StartDate]' AND '[EndDate]'

UNION ALL

-- 9. Invoice Posted (Change from Parked to Posted)
SELECT
    CONCAT(B.BUKRS, B.BELNR, B.GJAHR) AS InvoiceNumber, 'Invoice Posted' AS Activity, TO_TIMESTAMP(H.UDATE, H.UTIME) AS EventTime, H.USERNAME AS ProcessorUser, L.NAME1 AS VendorName, B.BUKRS AS CompanyCode, B.WAERS AS Currency, (SELECT SUM(WRBTR) FROM BSEG S WHERE S.BELNR = B.BELNR AND S.GJAHR = B.GJAHR AND S.BUKRS = B.BUKRS) AS InvoiceGrossAmount, S1.ZFBDT AS InvoiceDueDate, S1.ZTERM AS PaymentTerms, S1.EBELN AS PurchaseOrderNumber, S1.ZLSPR AS PaymentBlockReason, '[Workflow Approver]' AS Approver
FROM CDHDR H JOIN CDPOS P ON H.CHANGENR = P.CHANGENR JOIN BKPF B ON SUBSTRING(H.OBJECTID, 5, 10) = B.BELNR AND SUBSTRING(H.OBJECTID, 1, 4) = B.BUKRS AND SUBSTRING(H.OBJECTID, 15, 4) = B.GJAHR JOIN BSEG AS S1 ON B.BELNR = S1.BELNR AND B.GJAHR = S1.GJAHR AND B.BUKRS = S1.BUKRS AND S1.KOART = 'K' JOIN LFA1 AS L ON B.LIFNR = L.LIFNR
WHERE H.OBJECTCLAS = 'BELEG' AND P.TABNAME = 'BKPF' AND P.FNAME = 'BSTAT' AND P.VALUE_OLD = 'V' AND P.VALUE_NEW = ''
AND B.BUKRS = '[YourCompanyCode]' AND H.UDATE BETWEEN '[StartDate]' AND '[EndDate]'

UNION ALL

-- 10. Due Date Passed Without Payment (Calculated)
SELECT 
    CONCAT(B.BUKRS, B.BELNR, B.GJAHR) AS InvoiceNumber, 'Due Date Passed Without Payment' AS Activity, TO_TIMESTAMP(S1.ZFBDT) AS EventTime, 'SYSTEM' AS ProcessorUser, L.NAME1 AS VendorName, B.BUKRS AS CompanyCode, B.WAERS AS Currency, (SELECT SUM(WRBTR) FROM BSEG S WHERE S.BELNR = B.BELNR AND S.GJAHR = B.GJAHR AND S.BUKRS = B.BUKRS) AS InvoiceGrossAmount, S1.ZFBDT AS InvoiceDueDate, S1.ZTERM AS PaymentTerms, S1.EBELN AS PurchaseOrderNumber, S1.ZLSPR AS PaymentBlockReason, '[Workflow Approver]' AS Approver
FROM BKPF AS B JOIN BSEG AS S1 ON B.BELNR = S1.BELNR AND B.GJAHR = S1.GJAHR AND B.BUKRS = S1.BUKRS AND S1.KOART = 'K' JOIN LFA1 AS L ON B.LIFNR = L.LIFNR
WHERE S1.AUGBL IS NULL AND S1.ZFBDT < CURRENT_DATE
AND B.BLART IN ('KR', 'RE') AND B.BUKRS = '[YourCompanyCode]' AND B.CPUDT BETWEEN '[StartDate]' AND '[EndDate]'

UNION ALL

-- 11. Payment Proposal Created
SELECT
    CONCAT(B.BUKRS, B.BELNR, B.GJAHR) AS InvoiceNumber, 'Payment Proposal Created' AS Activity, TO_TIMESTAMP(H.LAUFD) AS EventTime, H.USNAM AS ProcessorUser, L.NAME1 AS VendorName, B.BUKRS AS CompanyCode, B.WAERS AS Currency, (SELECT SUM(WRBTR) FROM BSEG S WHERE S.BELNR = B.BELNR AND S.GJAHR = B.GJAHR AND S.BUKRS = B.BUKRS) AS InvoiceGrossAmount, S1.ZFBDT AS InvoiceDueDate, S1.ZTERM AS PaymentTerms, S1.EBELN AS PurchaseOrderNumber, S1.ZLSPR AS PaymentBlockReason, '[Workflow Approver]' AS Approver
FROM REGUH H
JOIN REGUP P ON H.LAUFD = P.LAUFD AND H.LAUFI = P.LAUFI AND H.XVORL = P.XVORL
JOIN BKPF B ON P.BUKRS = B.BUKRS AND P.BELNR = B.BELNR AND P.GJAHR = B.GJAHR
JOIN BSEG AS S1 ON B.BELNR = S1.BELNR AND B.GJAHR = S1.GJAHR AND B.BUKRS = S1.BUKRS AND S1.KOART = 'K'
JOIN LFA1 AS L ON B.LIFNR = L.LIFNR
WHERE B.BUKRS = '[YourCompanyCode]' AND H.LAUFD BETWEEN '[StartDate]' AND '[EndDate]'

UNION ALL

-- 12. Invoice Cleared By Payment
SELECT
    CONCAT(B.BUKRS, B.BELNR, B.GJAHR) AS InvoiceNumber, 'Invoice Cleared By Payment' AS Activity, TO_TIMESTAMP(S1.AUGDT) AS EventTime, B.USNAM AS ProcessorUser, L.NAME1 AS VendorName, B.BUKRS AS CompanyCode, B.WAERS AS Currency, (SELECT SUM(WRBTR) FROM BSEG S WHERE S.BELNR = B.BELNR AND S.GJAHR = B.GJAHR AND S.BUKRS = B.BUKRS) AS InvoiceGrossAmount, S1.ZFBDT AS InvoiceDueDate, S1.ZTERM AS PaymentTerms, S1.EBELN AS PurchaseOrderNumber, '' AS PaymentBlockReason, '[Workflow Approver]' AS Approver
FROM BKPF AS B
JOIN BSEG AS S1 ON B.BELNR = S1.BELNR AND B.GJAHR = S1.GJAHR AND B.BUKRS = S1.BUKRS AND S1.KOART = 'K'
JOIN LFA1 AS L ON B.LIFNR = L.LIFNR
WHERE S1.AUGBL IS NOT NULL
AND B.BLART IN ('KR', 'RE') AND B.BUKRS = '[YourCompanyCode]' AND S1.AUGDT BETWEEN '[StartDate]' AND '[EndDate]'

UNION ALL

-- 13. Invoice Reversed
SELECT
    CONCAT(B.BUKRS, B.BELNR, B.GJAHR) AS InvoiceNumber, 'Invoice Reversed' AS Activity, TO_TIMESTAMP(B.STODT) AS EventTime, B.USNAM_REV AS ProcessorUser, L.NAME1 AS VendorName, B.BUKRS AS CompanyCode, B.WAERS AS Currency, (SELECT SUM(WRBTR) FROM BSEG S WHERE S.BELNR = B.BELNR AND S.GJAHR = B.GJAHR AND S.BUKRS = B.BUKRS) AS InvoiceGrossAmount, S1.ZFBDT AS InvoiceDueDate, S1.ZTERM AS PaymentTerms, S1.EBELN AS PurchaseOrderNumber, S1.ZLSPR AS PaymentBlockReason, '[Workflow Approver]' AS Approver
FROM BKPF AS B
JOIN BSEG AS S1 ON B.BELNR = S1.BELNR AND B.GJAHR = S1.GJAHR AND B.BUKRS = S1.BUKRS AND S1.KOART = 'K'
JOIN LFA1 AS L ON B.LIFNR = L.LIFNR
WHERE B.STBLG IS NOT NULL
AND B.BLART IN ('KR', 'RE') AND B.BUKRS = '[YourCompanyCode]' AND B.STODT BETWEEN '[StartDate]' AND '[EndDate]'
`

数据模板：应付账款发票处理

您的应付账款发票处理数据模板

应付账款发票处理属性

应付账款发票处理活动

提取指南

用于应付账款发票数据提取的定制化ABAP程序

步骤

配置

a 查询示例 abap

直接对SAP ECC底层数据库执行SQL查询

步骤

配置

a 查询示例 sql

通过SAP RFC协议进行Python脚本提取

步骤

配置

a 查询示例 python

带有认证 SAP 连接器的 ETL 工具（例如：Theobald）

步骤

配置

a 查询示例 config

准备好开始了吗？

杜绝滞纳金：立即优化应付账款，节省60%成本！