まず影響の大きさと不確実性の高さで要因を整理し、影響大かつ不確実性大の二つを軸に選びます。例えば「幹線道路での完全無人運転の規制承認の速さ」と「ドライバー不足の深刻度」です。二軸で四つのシナリオを描き、それぞれで自社の売上構造、必要な資産、競合の動きを具体的に記述します。次に各シナリオで有効な打ち手を列挙し、全シナリオに共通して有効なもの（配車データの整備、ドライバーの多能工化、車両を長期保有せずリースで柔軟に入れ替える等）をロバスト戦略として優先します。シナリオ固有の打ち手は先行指標を定めて段階的に実行します。First, sort the driving forces by impact and uncertainty, and choose the two with high impact and high uncertainty as axes, for example 'speed of regulatory approval for fully driverless highway operation' and 'severity of the driver shortage'. The two axes yield four scenarios; describe each in terms of revenue structure, required assets, and competitor moves. Next, list effective actions per scenario and identify those that work across all four, such as building dispatch data infrastructure, cross-training drivers, and leasing rather than owning vehicles for flexibility. Prioritize these as the robust strategy. Scenario-specific actions are tied to early-warning indicators and executed in stages.
Peter Schwartz の『シナリオ・プランニングの技法』や GBN の手法では、「影響度×不確実性」のマトリクスで軸を選ぶことが標準です。確実性の高い要因（人口動態など）は全シナリオに共通する前提として扱い、軸にはしません。四つのシナリオは単なる楽観・悲観ではなく、軸の組み合わせによって質的に異なる世界として描く必要があります。ロバスト戦略の抽出は、各シナリオで打ち手の有効性を評価する「ウィンドタンネル」と呼ばれる手法で行われます。物流業界では、自動運転の技術成熟だけでなく、労働規制、保険制度、荷主の受容といった非技術要因が普及速度を左右するため、軸の選定段階で多様な専門家の意見を取り入れることが重要です。Peter Schwartz's The Art of the Long View and the Global Business Network method standardize axis selection through an impact-versus-uncertainty matrix. Highly certain forces such as demographics are treated as common assumptions, not axes. The four scenarios should be qualitatively different worlds arising from combinations of the axes, not merely optimistic and pessimistic versions. Robust strategies are extracted through 'wind-tunneling', in which each candidate action is tested for effectiveness under every scenario. In logistics, non-technical factors such as labor regulation, insurance regimes, and shipper acceptance shape adoption speed as much as technical maturity, so bringing in diverse experts during axis selection is essential.
軸の選び方次第でシナリオの内容が大きく変わるため、担当者の思い込みが結果に反映されやすい点が課題です。またロバスト戦略は「どの未来でも損をしない」代わりに、特定の未来で大きく勝つ機会を逃す平均的な戦略になりがちです。Because the choice of axes largely determines the scenarios, the planner's assumptions easily shape the outcome. Robust strategies also tend to be average strategies: they avoid losing in any future but may forgo the chance to win big in a specific one. A ten-year horizon also invites technological surprises that no axis anticipated.
軸の候補を複数組み合わせて試し、外部の専門家や現場ドライバーも交えてレビューします。ロバスト戦略を土台にしつつ、特定シナリオに賭ける「オプション投資」を小規模に併用し、先行指標が動いた時点で追加投資を判断する二層構造にします。Test several combinations of candidate axes and review them with outside experts and frontline drivers. Build a two-layer strategy: a robust foundation plus small option-like bets on specific scenarios, with additional investment triggered when early-warning indicators move. Revisit the axes themselves every two to three years so the framework does not ossify around yesterday's uncertainties.