# 第 24317 題：有限流網路 [返回所在分頁 122](https://ocme.evemisslab.com/zh-Hant/questions/page/122/) [← 逐題詳解目錄](https://ocme.evemisslab.com/zh-Hant/questions/) · [原題批次](https://ocme.evemisslab.com/zh-Hant/practice/batch-20261015-advance-20261113-r07-finite-flow-network-foundations/#q-20261015-advance-20261113-r07-finite-flow-network-0042) 本頁是已完成第二對話 AI 數理與中英語義覆核的教學詳解，只支持本題明示的模型與前提。沒有人工審查或新 Lean 證明，也不提升原題或 MKO 的 Canonical 真值狀態。 R0047-24317 · q-20261015-advance-20261113-r07-finite-flow-network-0042 ## 原題與已知條件 finite loop-free directed network、逐edge capacity與逐internal-vertex conservation。操作=flow-capacity-audit；資料={"vertices":["s16","a16","t16","u16"],"edges":[["s16","a16"],["a16","s16"],["a16","t16"],["t16","a16"]],"source":"s16","sink":"t16","capacities":[85,85,85,85],"flows":[102,34,51,17]}。請依 exact_ordered_tuple_declared_by_interface 欄序回答。 - G001 — finite loop-free directed network、逐edge capacity與逐internal-vertex conservation。操作=flow-capacity-audit；資料={"vertices":["s16","a16","t16","u16"],"edges":[["s16","a16"],["a16","s16"],["a16","t16"],["t16","a16"]],"source":"s16","sink":"t16","capacities":[85,85,85,85],"flows":[102,34,51,17]}。請依 exact_ordered_tuple_declared_by_interface 欄序回答。 ## 定義與適用範圍 沿用原題明示的論域、假設與來源 MKO 定義。本題已完成第二對話數理與中英語義覆核，並由原始參數逐題獨立重算模型、答案與每個算術步驟；原題仍為 candidate。 - [mko-finite-flow-network](https://ocme.evemisslab.com/zh-Hant/mko/mko-finite-flow-network/) ## 逐步推導 S001 · implies操作前 source roles = [flow-capacity-audit] 操作後 derived = [[source, sink, edge_capacity_flow_rows, capacity_rows, capacity_feasible, first_failed_edge_or_null, certificate]] 條件：無 本頁列完整來源欄序，不需JavaScript；gradient boundary依原mode選6／8／7欄，不補欄也不由答案猜mode。 步驟識別與依賴規則：source_ordered_fields_without_JS · 依賴：G001 S002 · implies操作前 source roles = [[s16, a16, t16, u16], [[s16, a16], [a16, s16], [a16, t16], [t16, a16]], [85, 85, 85, 85], [102, 34, 51, 17], s16, t16] 操作後 derived = [[a16, s16, t16, u16], [[[a16, s16], 85, 34], [[a16, t16], 85, 51], [[s16, a16], 85, 102], [[t16, a16], 85, 17]], 2, 是] 條件：無 先依原index配好edge／capacity／flow，再按來源English canonical(edge)序排；vertices用UTF16。反向邊是不同edge，zero容量與孤立點合法，負flow保留供容量判失敗，不修改。 步驟識別與依賴規則：Flow_zip_source_edges_capacities_flows_BEFORE_order · 依賴：G001, S001 S003 · implies操作前 source roles = [0, [a16, s16], 34, 85] 操作後 derived = [是, 是, 是] 條件：無 每條有向邊都要0<=flow<=capacity；網路signature合格並不保證候選flow可行。 步驟識別與依賴規則：Flow_EACH_edge_two_sided_capacity_obligation · 依賴：G001, S002 S004 · implies操作前 source roles = [1, [a16, t16], 51, 85] 操作後 derived = [是, 是, 是] 條件：無 每條有向邊都要0<=flow<=capacity；網路signature合格並不保證候選flow可行。 步驟識別與依賴規則：Flow_EACH_edge_two_sided_capacity_obligation · 依賴：G001, S003 S005 · implies操作前 source roles = [2, [s16, a16], 102, 85] 操作後 derived = [是, 否, 否] 條件：無 每條有向邊都要0<=flow<=capacity；網路signature合格並不保證候選flow可行。 步驟識別與依賴規則：Flow_EACH_edge_two_sided_capacity_obligation · 依賴：G001, S004 S006 · implies操作前 source roles = [3, [t16, a16], 17, 85] 操作後 derived = [是, 是, 是] 條件：無 每條有向邊都要0<=flow<=capacity；網路signature合格並不保證候選flow可行。 步驟識別與依賴規則：Flow_EACH_edge_two_sided_capacity_obligation · 依賴：G001, S005 S007 · calculation操作前 Flow ALL incoming at vertex a16 := sum(102, 17) 操作後 Flow ALL incoming at vertex a16 = 119 條件：無 依原題角色與已證中間量精確代值；分母先核非零，不以近似數替代。 步驟識別與依賴規則：exact_rational_arithmetic · 依賴：G001, S006 S008 · calculation操作前 Flow ALL outgoing at vertex a16 := sum(34, 51) 操作後 Flow ALL outgoing at vertex a16 = 85 條件：無 依原題角色與已證中間量精確代值；分母先核非零，不以近似數替代。 步驟識別與依賴規則：exact_rational_arithmetic · 依賴：G001, S007 S009 · implies操作前 source roles = [a16, [2, 3], [0, 1]] 操作後 derived = [119, 85, 否, 是] 條件：無 入邊／出邊各依完整清單加總；孤立點是0=0。守恆只要求internal vertices，源匯仍計完整balance以求net值。 步驟識別與依賴規則：Flow_EACH_vertex_complete_directional_incidence · 依賴：G001, S008 S010 · calculation操作前 Flow ALL incoming at vertex s16 := sum(34) 操作後 Flow ALL incoming at vertex s16 = 34 條件：無 依原題角色與已證中間量精確代值；分母先核非零，不以近似數替代。 步驟識別與依賴規則：exact_rational_arithmetic · 依賴：G001, S009 S011 · calculation操作前 Flow ALL outgoing at vertex s16 := sum(102) 操作後 Flow ALL outgoing at vertex s16 = 102 條件：無 依原題角色與已證中間量精確代值；分母先核非零，不以近似數替代。 步驟識別與依賴規則：exact_rational_arithmetic · 依賴：G001, S010 S012 · implies操作前 source roles = [s16, [0], [2]] 操作後 derived = [34, 102, 否, 否] 條件：無 入邊／出邊各依完整清單加總；孤立點是0=0。守恆只要求internal vertices，源匯仍計完整balance以求net值。 步驟識別與依賴規則：Flow_EACH_vertex_complete_directional_incidence · 依賴：G001, S011 S013 · calculation操作前 Flow ALL incoming at vertex t16 := sum(51) 操作後 Flow ALL incoming at vertex t16 = 51 條件：無 依原題角色與已證中間量精確代值；分母先核非零，不以近似數替代。 步驟識別與依賴規則：exact_rational_arithmetic · 依賴：G001, S012 S014 · calculation操作前 Flow ALL outgoing at vertex t16 := sum(17) 操作後 Flow ALL outgoing at vertex t16 = 17 條件：無 依原題角色與已證中間量精確代值；分母先核非零，不以近似數替代。 步驟識別與依賴規則：exact_rational_arithmetic · 依賴：G001, S013 S015 · implies操作前 source roles = [t16, [1], [3]] 操作後 derived = [51, 17, 否, 否] 條件：無 入邊／出邊各依完整清單加總；孤立點是0=0。守恆只要求internal vertices，源匯仍計完整balance以求net值。 步驟識別與依賴規則：Flow_EACH_vertex_complete_directional_incidence · 依賴：G001, S014 S016 · calculation操作前 Flow ALL incoming at vertex u16 := sum() 操作後 Flow ALL incoming at vertex u16 = 0 條件：無 依原題角色與已證中間量精確代值；分母先核非零，不以近似數替代。 步驟識別與依賴規則：exact_rational_arithmetic · 依賴：G001, S015 S017 · calculation操作前 Flow ALL outgoing at vertex u16 := sum() 操作後 Flow ALL outgoing at vertex u16 = 0 條件：無 依原題角色與已證中間量精確代值；分母先核非零，不以近似數替代。 步驟識別與依賴規則：exact_rational_arithmetic · 依賴：G001, S016 S018 · implies操作前 source roles = [u16, [], []] 操作後 derived = [0, 0, 是, 是] 條件：無 入邊／出邊各依完整清單加總；孤立點是0=0。守恆只要求internal vertices，源匯仍計完整balance以求net值。 步驟識別與依賴規則：Flow_EACH_vertex_complete_directional_incidence · 依賴：G001, S017 S019 · calculation操作前 Flow source NET outflow not gross outgoing := sub(102, 34) 操作後 Flow source NET outflow not gross outgoing = 68 條件：無 依原題角色與已證中間量精確代值；分母先核非零，不以近似數替代。 步驟識別與依賴規則：exact_rational_arithmetic · 依賴：G001, S018 S020 · calculation操作前 Flow sink NET inflow not gross incoming := sub(51, 17) 操作後 Flow sink NET inflow not gross incoming = 34 條件：無 依原題角色與已證中間量精確代值；分母先核非零，不以近似數替代。 步驟識別與依賴規則：exact_rational_arithmetic · 依賴：G001, S019 S021 · implies操作前 source roles = [s16, t16, [[[a16, s16], 34, 85, 是], [[a16, t16], 51, 85, 是], [[s16, a16], 102, 85, 否], [[t16, a16], 17, 85, 是]], [[a16, 119, 85, 否], [u16, 0, 0, 是]], 68, 34] 操作後 derived = [否, 否, 否, 否, 否] 條件：無 全vertex總in−out中每edge相消，故全部internal守恆必得兩net相等；反向不成立，internal imbalance可互抵。容量、逐點守恆、兩net值分別核，feasible不推出maximum/min-cut。最後No是不提供最優證書，不否定特例恰好最優。 步驟識別與依賴規則：Flow_value_capacity_conservation_and_maximum_separate · 依賴：G001, S020 ## 逐步反向核對 - S021 · premise_and_result_audit — 回到原始資料與本步條件核對，不把單向推論誤稱等價。 - S020 · premise_and_result_audit — 以精確有理數重算 Flow sink NET inflow not gross incoming，不以小數近似替代。 - S019 · premise_and_result_audit — 以精確有理數重算 Flow source NET outflow not gross outgoing，不以小數近似替代。 - S018 · premise_and_result_audit — 回到原始資料與本步條件核對，不把單向推論誤稱等價。 - S017 · premise_and_result_audit — 以精確有理數重算 Flow ALL outgoing at vertex u16，不以小數近似替代。 - S016 · premise_and_result_audit — 以精確有理數重算 Flow ALL incoming at vertex u16，不以小數近似替代。 - S015 · premise_and_result_audit — 回到原始資料與本步條件核對，不把單向推論誤稱等價。 - S014 · premise_and_result_audit — 以精確有理數重算 Flow ALL outgoing at vertex t16，不以小數近似替代。 - S013 · premise_and_result_audit — 以精確有理數重算 Flow ALL incoming at vertex t16，不以小數近似替代。 - S012 · premise_and_result_audit — 回到原始資料與本步條件核對，不把單向推論誤稱等價。 - S011 · premise_and_result_audit — 以精確有理數重算 Flow ALL outgoing at vertex s16，不以小數近似替代。 - S010 · premise_and_result_audit — 以精確有理數重算 Flow ALL incoming at vertex s16，不以小數近似替代。 - S009 · premise_and_result_audit — 回到原始資料與本步條件核對，不把單向推論誤稱等價。 - S008 · premise_and_result_audit — 以精確有理數重算 Flow ALL outgoing at vertex a16，不以小數近似替代。 - S007 · premise_and_result_audit — 以精確有理數重算 Flow ALL incoming at vertex a16，不以小數近似替代。 - S006 · premise_and_result_audit — 回到原始資料與本步條件核對，不把單向推論誤稱等價。 - S005 · premise_and_result_audit — 回到原始資料與本步條件核對，不把單向推論誤稱等價。 - S004 · premise_and_result_audit — 回到原始資料與本步條件核對，不把單向推論誤稱等價。 - S003 · premise_and_result_audit — 回到原始資料與本步條件核對，不把單向推論誤稱等價。 - S002 · premise_and_result_audit — 回到原始資料與本步條件核對，不把單向推論誤稱等價。 - S001 · premise_and_result_audit — 回到原始資料與本步條件核對，不把單向推論誤稱等價。 ## 結論 (s16, t16, [[[a16, s16], 85, 34], [[a16, t16], 85, 51], [[s16, a16], 85, 102], [[t16, a16], 85, 17]], [[[a16, s16], 34, 85, 是], [[a16, t16], 51, 85, 是], [[s16, a16], 102, 85, 否], [[t16, a16], 17, 85, 是]], 否, [[s16, a16], 102, 85, 否], every_edge_requires_zero_less_equal_flow_less_equal_capacity) 依上列完整代值及條件核對得到原題答案；原題編號、格式與狀態保持不變。 ## 本題覆核與來源綁定 覆核日期：2026-10-26 原題、答案、步驟表達式、數值 AST 與既有歷史保持原樣；原來的候選狀態屬於來源歷史，不覆寫為形式化證明。 來源雜湊與覆核憑證 - input commit: 0f7be61e2c22317a4075561dc822c87d881a50db - original question SHA-256: 8ae10d2bc097bd90dc558eddec62926ce4853754325f27232cc5032155ff6650 - candidate record SHA-256: 0ea9b7351f32812060a7b2ce041916390d988adc8d1837b72c0c6b2acdca4532 - semantic review receipt SHA-256: 9b43b9823e9bd2478d6caa5b57ffed161c2c8956af86974e901d5662ad994386 - question-first witness SHA-256: eb97596068e2d5648033bab2b8f6ca9b653bbe8cb77c0f8e1d16d0c40e46f50b - Lean: NOT_RUN · human review: NOT_PERFORMED [同源完整 JSON](https://ocme.evemisslab.com/data/question-derivations/questions/q-20261015-advance-20261113-r07-finite-flow-network-0042.json)