highest-cost-scenario-identification

# Highest Cost Scenario Identification This skill covers two question patterns in the dabstep dataset, both requiring you to find the scenario that maximizes the payment processing fee. ## Fee Formula From `manual.md`: ``` fee = fixed_amount + rate * transaction_value / 10000 ``` All fee rules are in `fees.json`. Key rule fields: - `card_scheme`: card network name (e.g. "GlobalCard", "NexPay", "TransactPlus", "SwiftCharge") - `is_credit`: `true` (credit only), `false` (debit only), `null` (applies to both) - `aci`: list of ACI letters this rule covers; `null` means all ACIs; **empty list `[]` means no ACI matches** (rule is ACI-agnostic/not ACI-specific — treat as applying to all ACIs when empty) - `merchant_category_code`: list of integer MCC codes this rule covers - `fixed_amount`, `rate`: fee components **Critical null semantics**: `null` in any field = applies to all values of that field. --- ## Question Type 1: Most Expensive MCC **Pattern**: "What is the most expensive MCC for a transaction of X euros, in general?" ### Algorithm ```python import json with open('fees.json') as f: fees = json.load(f) transaction_value = X # euros, from the question mcc_max_fee = {} # mcc_code -> max fee seen for rule in fees: fee = rule['fixed_amount'] + rule['rate'] * transaction_value / 10000 for mcc in rule['merchant_category_code']: if mcc not in mcc_max_fee or fee > mcc_max_fee[mcc]: mcc_max_fee[mcc] = fee overall_max = max(mcc_max_fee.values()) result_mccs = sorted([mcc for mcc, f in mcc_max_fee.items() if f == overall_max]) print(', '.join(str(m) for m in result_mccs)) ``` ### Critical Rules - **Do NOT filter MCCs by `merchant_category_codes.csv`**. Use all MCCs that appear in `fees.json` rules. The fees.json is the authoritative source for which MCCs have fee rules. - "In general" means: find the maximum possible fee over ALL rules, with no restriction on card_scheme, is_credit, aci, account_type, capture_delay, etc. - If multiple MCCs tie at the maximum fee, list all of them sorted numerically. ### Common Error to Avoid Wrong: Filtering MCCs to only those in `merchant_category_codes.csv` (this incorrectly excludes valid MCCs like 3003, 7231 that appear in fees.json but not in the MCC reference file). --- ## Question Type 2: Most Expensive ACI for a Specific Card Scheme **Pattern**: "For a credit transaction of X euros on [CardScheme], what would be the most expensive ACI? In the case of a draw, return the ACI with the lowest alphabetical order." ### Algorithm ```python import json with open('fees.json') as f: fees = json.load(f) card_scheme = "GlobalCard" # from the question transaction_value = 1.0 # euros, from the question ALL_ACIS = ['A', 'B', 'C', 'D', 'E', 'F', 'G'] aci_max_fee = {} # aci -> max fee for rule in fees: # Filter by card scheme if rule['card_scheme'] != card_scheme: continue # Filter by credit: keep if is_credit is True or None (null = applies to both) if rule['is_credit'] == False: continue fee = rule['fixed_amount'] + rule['rate'] * transaction_value / 10000 # Determine which ACIs this rule covers rule_acis = rule['aci'] if rule_acis is None: rule_acis = ALL_ACIS # null = all ACIs elif len(rule_acis) == 0: rule_acis = ALL_ACIS # empty list = applies to all ACIs (no ACI restriction) for aci in rule_acis: if aci not in aci_max_fee or fee > aci_max_fee[aci]: aci_max_fee[aci] = fee # Find the ACI(s) with the highest fee if not aci_max_fee: print("Not Applicable") else: max_fee = max(aci_max_fee.values()) best_acis = sorted([a for a, f in aci_max_fee.items() if f == max_fee]) # Tiebreak: alphabetically lowest answer = best_acis[0] print(f"['{answer}']") ``` ### Critical Rules - **Filter `is_credit`**: For "credit transaction" questions, keep rules where `is_credit is True` or `is_credit is None`. Exclude rules where `is_credit == False`. - **ACI null vs empty list**: Both `null` and `[]` in the `aci` field mean the rule applies to all ACIs. Do NOT skip rules with empty aci lists. - **Output format**: The answer is the ACI **letter** (e.g. `C`, `E`), NOT the fee value. Return in list format: `['C']`. - **Tiebreak**: If multiple ACIs share the maximum fee, return the one that comes first alphabetically. ### Common Errors to Avoid 1. **Outputting the fee amount** (e.g. "0.23") instead of the ACI letter — the question asks WHICH ACI, not what the fee is. 2. **Treating empty `aci` list as "no ACIs"** — empty list means all ACIs apply (same as null). 3. **Forgetting null is_credit rules** — rules with `is_credit: null` apply to credit transactions too. --- ## Output Format - MCC questions: comma-separated list of integer MCC codes, sorted numerically. E.g.: `3000, 3001, 3002, 3003, 7011, 7032, 7512, 7513` - ACI questions: single letter in a list. E.g.: `['C']` - If no applicable rules exist: `Not Applicable` ## Validation Checklist Before finalizing your answer: 1. Did you use the correct fee formula: `fixed_amount + rate * transaction_value / 10000`? 2. For MCC questions: did you search ALL fee rules without filtering by card scheme or credit type? 3. For ACI questions: did you filter by the correct card scheme AND include `is_credit=null` rules? 4. Is your output the MCC codes / ACI letters (not the fee amount)? 5. Are ties handled correctly (all MCCs listed / alphabetically first ACI)?